The Externality of Population Growth

Do We Owe the Future More Children?

On the long, agonising flight home from SF, I finally read Dean Spears and Michael Geruso’s After the Spike: Population, Progress, and the Case for People. It had been on my list since the start of the year, when I became increasingly interested in Singapore’s falling birth rate.

The book’s warning is pretty simple: on the world’s current trajectory, population will peak and begin to decline before the end of this century. Because demography moves slowly, waiting until the consequences are obvious may mean waiting until they are difficult to reverse.

Spears and Geruso make a case for eventual population stability. They argue that more people mean more ideas, deeper specialisation and more shoulders across which to spread the fixed costs of institutions and technological progress. They are also clear that their answer is not coercion, but a reorganisation of society so that people who want children can have them without surrendering their other aspirations. And they definitely do not claim to have all the answers on how to realistically do this. One side argument is also their insistence that depopulation is a poor substitute for environmental policy; instead, climate change should be addressed directly through decarbonisation, regulation and technological progress.

It is a provocative book, and I recommend it to anyone who cares about what population growth means for our collective lives (particularly to policymakers). But for me, the question that stayed was not whether depopulation is coming; instead it was whether a child can really be understood as an economic externality.

That framing highlights something important, but also risks obscuring something more important still.

The seduction of the externality and climate analogy

Depopulation, like climate change, may emerge from millions of individually reasonable choices that accumulate into a future nobody chose.

A person deciding whether to have another child thinks about the costs, sacrifices and possibilities of their own life. They do no, and probably should not, calculate what that decision will do to the taxation structure, the healthcare industry or humanity’s innovative capacity decades after they are gone.

Economists call this an externality: a cost or benefit borne by someone outside the original private decision. In the population case, the externality is also uniquely inter-temporal. Parents bear much of the cost of raising children now, but many of the alleged benefits arrive later, when those children become workers, caregivers, taxpayers, inventors or parents themselves. In essence, the people who paid the initial cost capture only a fraction of the social return.

It’s understandable why the comparison with climate change is immediately appealing. It can be privately rational for one person to take a flight, drive to work or leave the air conditioning on, and so the climate problem appears when billions of emissions become collectively externally detrimental.

Population has an additional inertia. Even if fertility rose tomorrow, babies would not immediately become workers or parents. By the time the effects of low fertility are unmistakable, the decisions that produced them may be decades old.

Viewing the population problem through an externality lens also tells you why politics compounds the problem. Elected governments are rewarded for solving visible problems within an electoral cycle, not for preventing a demographic imbalance whose costs will be borne by people who cannot yet vote.

So in all, the externality lens helps us see why individually free choices and collectively desirable outcomes need not align.

Some claims and their certainties

The case against carbon emissions as an externality begins with a well-established causal mechanism. Greenhouse gases warm the planet, and that warming imposes harms. The magnitude and distribution of those harms are often debated, but the direction of the effect is as scientific as it gets,

On the other hand, calling population growth a positive externality actually bundles together three different claims.

The first is arithmetic and rather uncontroversial. If fertility stays below replacement, successive generations shrink and population eventually declines.

The second is a forecast. Fertility will remain below replacement long enough to produce sustained global depopulation. This is increasingly plausible, but it remains a prediction about social behaviour rather than a law. The UN’s latest central projection puts the global population peak at roughly 10.3 billion in the mid-2080s and estimates an 80% chance that the peak will occur this century. Spears and Geruso add that countries whose completed fertility has fallen substantially below replacement have not subsequently restored it. For now, let’s entertain this forecast as fact.

The third claim is slightly more contestable: a smaller population will necessarily make humanity worse off.

Rapid ageing can strain pensions and healthcare systems, and a shrinking workforce can destabilise institutions built around continuous growth. But the severity of those problems depends on so many factors such as productivity, automation, immigration, retirement ages, healthy lifespans and political choices. Demography sets constraints for paths of economic development but it does not dictate a single economic future.

In other words, there is considerably more unresolved internal economics in the population externality than in the carbon one.

More people, or more human potential?

The book’s strongest economic argument is that more people mean more ideas.

If you studied Economics in university, you would easily understand why: modern growth theory treats ideas as non-rivalrous. If I eat a loaf of bread, you cannot eat the same loaf. But if someone discovers a useful technology, millions of people can use it without exhausting it. A larger population can produce more researchers, more combinations of knowledge and more discoveries from which everyone benefits.

Yet I would argue that population size is not the same thing as innovative capacity.

More recent research (for example by Nicholas Bloom, Charles Jones, John Van Reenen and Michael Webb) suggests that ideas are becoming harder to find: across several fields, sustaining a similar rate of progress has required sharply increasing research effort. In their framework, growth depends not only on the number of researchers but also on how productive their research is, which in turn is a function of other exogenous factors, often institutional and political.

Nor does every person have an equal chance to become an effective researcher. Innovation depends on many other factors like education, health, nutrition, institutions, funding, mentorship, laboratories, professional networks and credible routes from discovery to application. Humanity may be surrounded by unrealised genius not simply because there are too few people, but because too many people never receive the conditions in which their abilities can develop.

It could therefore be argued that a society can therefore expand its effective population of innovators without expanding its total population. In certain countries, this can entail removing barriers faced by marginalised groups (women, the poor) and generally people born far from centres of opportunity. In most countries, this means improving our education (especially in the age of AI) and public health (from a productivity, capacity perspective). Increasingly, this can also entail using AI to extend what each citizen is capable of doing. What remains then is to estimate how far away we are from realising this potential given our current state of innovation (and population), before discarding this pursuit as less impactful than that of relying on population growth.

To be clear, this does not refute the book’s case. All else equal, a larger population may (and probably will) still produce more ideas. But all else is never equal, and a billion additional people whose potential is neglected may contribute less to progress than a fraction of that number living in institutions that allow them to flourish.

One additional consideration is that the sign of the population externality may also change with scale. In a small or rapidly shrinking society, another person’s contribution to specialisation, care and institutional resilience may be especially valuable. In a large society pressing against environmental or infrastructural limits, the marginal costs may loom larger. Just look at the densest urban cities in the world and you will know what I’m referring to. Quality of life is a moving target with short-term memory, and even though we are all many times better-off than our ancestors from a century ago, you would still find many in such urban areas wanting of what they would consider a high-quality life.

The uncomfortable reality we have to ultimately grapple with is that we have no credible calculation of the optimum. Spears and Geruso favour stabilisation: roughly replacement fertility and zero long-run population growth. That may be a sensible political destination, but it is not an optimum derived from their economic logic. If more people create benefits through ideas, specialisation and scale, why should those benefits stop precisely at zero growth?

Their answer seems pragmatic rather than theoretical. Restoring even replacement fertility may be extraordinarily difficult. Stabilisation is already ambitious. But that is different from proving it is the ideal population path. In my opinion, the authors would have had a more compelling case if they had kept the normative separate from the empirical.

A person is not the same kind of economic thing as a tonne of carbon

The deeper problem with the externality analogy is in fact moral and not empirical.

A tonne of carbon is an unwanted residue of an otherwise useful activity, and its external effect has a clear direction. However, a person does not have the same property.

An additional person consumes land and energy, but also works, creates demand and cares for others. They draw on public institutions and later sustain them. They may create knowledge or help solve the problems their existence contributes to.

Most importantly, the experience of life has a value cannot be reduced to its effect on GDP, pension solvency or the government balance sheet.

The external effect of a person depends on how that person lives and, crucially, on how society is organised around them. A tonne of carbon warms the atmosphere regardless of whether it was emitted in a rich country or a poor one. A child, by contrast, does not arrive as a ready-made worker, taxpayer or inventor. Whether they become any of those things depends on the education, health, freedoms and opportunities available to them.

This matters because demographic anxiety can make a headcount solution seem sufficient for what may really be an institutional problem. If fewer workers must support more retirees, one response is to produce more future workers. But societies can also raise productivity, automate routine work, redesign care, integrate migrants, extend healthy working lives and make better use of people whose talents are currently excluded or wasted.

None of these is a perfect substitute for a younger population, but neither are additional births a substitute for well-functioning institutions. A society that fails to educate its children, enable meaningful labour participation in the economy or translate technology into broadly shared prosperity will not solve those failures merely by having more people. Conversely, a smaller but more productive and better-organised society may be more resilient than its raw population numbers suggest.

The supposed positive externality, then, does not reside in the birth alone. It emerges from the relationship between a person and the society that develops their capabilities. The relevant policy question is not simply how to create more future people, but how to enable each person to contribute more, and to live a high-quality life.

What society owes

We cannot derive the correct population size from science, nor should we treat people as instruments for meeting a demographic target. No externality argument can justify coercion, restrictions on reproductive autonomy or the suggestion that anyone owes society a child.

But it can expose a mismatch in how those institutions are organised. Raising children is treated largely as a private choice, so parents absorb much of the cost in time, income and forgone opportunities. Yet the resulting benefits of future workers, caregivers and citizens are shared widely. Childcare, parental leave, family-sized housing and protection from career penalties are ways of correcting that mismatch, not subsidies for manufacturing future taxpayers and innovators. They allow the cost of sustaining society to be shared more like the benefits already are.

Increasingly, I feel that the aim should not be to maximise births. It should be to narrow the gap between the children people want and the children they feel able to have. A humane population policy would ask present society to account for future benefits, without making any individual responsible for delivering them.

From an externality perspective, population decline is not simply analogous to climate change. The forecast is less certain, the externality has no fixed sign, and the consequences depend on productivity and institutions rather than headcount alone.

But the analogy does reveal a genuine institutional failure, specifically that we are good at registering present costs and bad at representing future dependence. Parents know exactly what raising a child demands today, but the future society that may benefit from their work cannot bargain with them in the present.

Anyway, After the Spike was a good read - I found it most useful not when it tells us how many people the future should contain, but when it asks whether the present has made enough room for that future to be freely chosen. The question is not whether people owe the future children, but it is whether a society that depends on future generations owes more to the people willing to raise them.