The Arithmetic of Age
How demographic decline is making China, America, and Europe older, calmer, more unequal, and less dynamic — and why immigration both postpones the reckoning and sharpens its politics
The slow variable beneath the fast politics
The preceding essays in this series argued that the United States and Britain are not yet plutocracies but sit in the antechamber of one — that wealth has concentrated to historic levels, that the legal machinery of protest, strike, and franchise has been narrowed, and that the organised countervailing power of ordinary people has hollowed out, so that the corrective mechanisms of democracy are weakening even as its forms remain intact. Those essays named a demographic engine driving the process but did not open the casing. This one does.
The claim here is that demography is the slow variable beneath the fast politics: the deep tide whose movement is nearly invisible on the timescale of an election but decisive on the timescale of a generation. Populations across the rich world, and now across much of the developing world, are ageing and in a growing number of cases shrinking. That single fact, worked through the mechanisms of political participation, economic behaviour, and public finance, produces a characteristic constellation — societies that are calmer in a specific and double-edged sense (fewer revolutions, a different profile of protest), more conservative in a way that is more subtle than the folk version admits, more unequal along axes that cut across the familiar rich-poor divide, and less economically dynamic. Immigration is the one great lever that alters this arithmetic, and it alters it in two opposite directions at once: it relieves the economic pressure of ageing while intensifying its politics.
Three cautions before the argument, because the standard of this series forbids overreach. First, none of what follows is demographic determinism. Demography sets probabilities and applies pressures; it does not dictate outcomes, and every mechanism described here is contingent, contested at its edges, and reversible in principle. Second, several of the popular claims in this territory — that people simply “get conservative with age,” that ageing straightforwardly kills economic growth, that immigration unambiguously depresses wages — are either false or seriously contested, and where the evidence cuts against the thesis it is presented in full rather than suppressed. Third, the countries compared here are at radically different points on the same curve, and the differences are as instructive as the trend. The argument is not that the world is becoming uniformly grey. It is that a common demographic process is producing recognisably similar political and economic effects across very different societies, and that understanding the process illuminates the politics of disparity that the earlier essays described.
The scale of the transition
Begin with the magnitudes, because they are larger and faster than most political commentary registers, and because the numbers here are anchored to the World Bank and UN Population Division rather than asserted.
The proximate driver is the collapse of fertility. A population replaces itself at a total fertility rate of about 2.1 children per woman. Almost no rich country is near it, and a growing list of developing countries have fallen below it. On the latest figures, the United States sits at roughly 1.6, the United Kingdom at 1.55, Germany at 1.36, Italy at 1.18, and Japan at 1.15. These are not transient dips; they are the settled condition of a generation. The extreme case is South Korea, whose total fertility rate has fallen to about 0.75 — the lowest ever recorded for a national population, and a figure that implies each generation will be less than half the size of the one before it if sustained. The pattern is no longer confined to the wealthy West and East Asia: India, long imagined as the archetype of a young and teeming society, has fallen to roughly 1.96, below replacement, while only sub-Saharan Africa — Nigeria still at about 4.4 — retains the high fertility that was universal within living memory.
The consequence, worked through with a lag, is ageing, and the speed of it in the latecomer countries is without historical precedent. The share of the population aged sixty-five and over now stands at about 18% in the United States, 20% in the United Kingdom, 24% in Germany, 25% in Italy, and 30% in Japan — the last meaning that nearly one Japanese person in three has passed the traditional retirement age. The old-age dependency ratio, the number of over-sixty-fives for every hundred people of working age, tells the story in the currency that matters for public finance: it has reached about 28 in the US, 31 in the UK, 38 in Germany, 40 in Italy, and an astonishing 51 in Japan, where there is now roughly one person over sixty-five for every two of working age. These societies are not ageing gently. They are inverting the age pyramid that every human society has taken for granted since the species began.
The single most consequential case is China, because of its scale and the peculiar violence of its trajectory. China’s fertility, already pushed down by the one-child policy enforced from 1979, has fallen to about 1.0 — comparable to the lowest in Europe. Its share of over-sixty-fives rose from about 7% around 2000 to roughly 15% today, crossing in barely two decades the threshold from “ageing” to “aged” society that took France more than a century and Japan about a quarter-century. Its population, having reached roughly 1.41 billion, peaked around 2021–2022 and has begun to fall. And it is doing all of this at a fraction of the income per head at which Japan, Germany, or the United States aged — the condition demographers summarise as growing old before growing rich (Eberstadt 2024; Greenhalgh 2008). China faces the fiscal and economic burdens of a rich-world age structure with a middle-income economy, and, as later sections argue, without the release valve that the Western democracies have used to soften the same pressure.
This is the tide. Everything that follows is an argument about what it does when it reaches the shore of politics and economics.
Why the young make revolutions and the old do not
Consider first the most dramatic of the effects, the one that connects demography to the sheer possibility of upheaval: revolutions and mass political violence are overwhelmingly the work of the young, and ageing societies lack the raw material to produce them.
The foundational modern statement is Jack Goldstone’s demographic-structural theory of revolution (Goldstone 1991). Studying the great upheavals of the early modern world, Goldstone argued that revolutions cluster in periods of rapid population growth that strain the capacity of states and elites to absorb the resulting pressures — fiscal crisis as governments struggle to fund themselves, intra-elite competition as the number of aspirants outruns the positions available, and, crucially, a swelling cohort of young people entering a labour market and a political order that cannot accommodate them. The mechanism is not that poverty causes revolution; it is that a particular age structure, combined with blocked opportunity, produces a large pool of people with the energy, the grievance, and the low opportunity cost required for high-risk collective action.
The quantitative literature has borne this out under the heading of the “youth bulge.” Henrik Urdal’s cross-national study found that societies with unusually large cohorts of young adults face a significantly elevated risk of political violence, particularly where economic opportunity and political voice are lacking (Urdal 2006). The intuition is straightforward and has been formalised in various ways: a young man without work, without property, and without prospects has little to lose from rebellion and a great deal of time and physical capacity to devote to it, and when such men are numerous relative to the established order’s ability to absorb them, the probability of organised violence rises. Gunnar Heinsohn pushed this into a stronger and more deterministic thesis linking youth bulges directly to war and conquest (Heinsohn 2003); that stronger version is genuinely controversial and has been criticised for overreach and for underweighting the political and economic conditions that determine whether a youth bulge becomes violent, and it should be cited as a provocation rather than an established finding. The defensible core, however, is robust: the supply of potential rebels is a function of age structure, and revolutionary and insurrectionary politics are disproportionately young.
Doug McAdam’s study of the 1964 Freedom Summer campaign supplies the individual-level mechanism through the concept of biographical availability (McAdam 1988). High-risk, high-commitment activism is concentrated among those free of the constraints — full-time careers, mortgages, dependent children, spousal obligations — that raise the personal cost of participation. The young are biographically available in a way the middle-aged and old, encumbered by exactly those commitments, are not. It is not that older people care less; it is that the structure of a settled adult life makes sustained, risky political engagement far more costly, and people respond to that cost.
Put the demographic transition and these mechanisms together and the implication is stark. The youth bulge is vanishing across the ageing world. Where the median society was once young, with a broad base of exactly the cohort that supplies revolutionary energy, the rich-world and now the East Asian and increasingly the South Asian societies are becoming old, with the demographic centre of gravity shifting to cohorts that are biographically encumbered, physically past the age of high-risk action, and, as the next sections argue, materially invested in the existing order. This does not mean ageing societies are politically placid — they can be angry, polarised, and volatile — but it does mean they are structurally unlikely to produce the classic revolutionary situation, because the demographic fuel for it is no longer there. The contrast with the youthful societies of sub-Saharan Africa, where the youth bulge is still growing, is not incidental; it is why the locus of revolutionary and insurrectionary risk is shifting geographically toward the last young regions of the earth even as the old regions become, in this specific sense, unshakeable.
Why the old protest less — and why “less” is the wrong word
The claim that older populations protest less requires care, because in its crude form it is contradicted by recent evidence, and the standard of this series is that inconvenient evidence is stated, not buried.
The crude form is that as a society ages, the streets go quiet. As a claim about the raw frequency of protest, this is false, and the earlier essays in this series said so: the 2010s and 2020s have been among the most protest-intensive decades in the modern history of the United States and Britain, with some of the largest single mobilisations either country has ever seen. An ageing society is manifestly capable of generating enormous protests. So the flat statement “older people protest less, therefore ageing societies protest less” does not survive contact with the data, and it should be abandoned in that form.
What survives, and what the evidence does support, operates at two levels that must be kept distinct. At the individual level, participation in non-institutional, disruptive, high-cost political action — occupying, marching, striking, direct action — is indeed concentrated among the young, for the biographical-availability reasons already given and because the young are the group whose grievances are least addressed by an order oriented toward those who already hold assets. Older citizens are more likely to participate through institutional channels — voting, donating, party membership where it survives — and less likely to be found in the high-risk repertoire. This is a robust and unsurprising finding of the political-participation literature.
At the aggregate level, the relevant effect is not that an ageing society protests less in total but that the composition and consequence of its politics shift. Three things follow. First, the demographic centre of gravity of the reliable electorate ages faster than the population, because older citizens vote at markedly higher rates, so the preferences politicians must court tilt toward the old. Second, the disruptive repertoire, being concentrated among a shrinking youth cohort, commands a shrinking share of the society’s political energy and can be more easily contained or ignored by an order whose decisive constituency lies elsewhere. Third — and this is the link to the earlier essays — the protest that does occur increasingly fails to translate into durable change, because the institutions that once converted mobilisation into countervailing power (above all organised labour) have collapsed, and because the median voter whose consent ultimately governs is drawn from the cohorts least sympathetic to disruption. The phenomenon the earlier essays called acquiescence is not silence. It is the decoupling of mobilisation from outcomes, and ageing sharpens that decoupling by shifting the decisive electorate toward stability and by thinning the youthful cohort in which disruptive energy concentrates. “Less protest” is the wrong phrase. “Protest that matters less” is the right one, and demography is one of the reasons.
Do people become conservative as they age?
Here lies the most misunderstood claim in the whole territory, and getting it right matters, because the popular version is largely wrong and the correct version points somewhere more interesting.
The folk claim is that individuals drift rightward as they age — that the radical of twenty becomes the reactionary of sixty as a matter of biological or psychological course. The scholarly evidence does not support this as the dominant effect. The bulk of what looks, in a cross-section, like “older people are more conservative” turns out on closer analysis to be a cohort effect rather than a life-cycle effect: different generations were formed in different political eras and carry those formations forward, so that at any moment the old differ from the young less because they have changed than because they were shaped by a different world and have largely kept the outlook they acquired when young.
The mechanism is the “impressionable years” hypothesis, for which the evidence is strong. Political attitudes and partisan identities crystallise in late adolescence and early adulthood and then become markedly more resistant to change; susceptibility to attitude change declines with age after the formative period (Krosnick and Alwin 1989; Sears and Funk 1999). The corollary is that a generation’s politics is stamped by the events of its youth and persists. Ghitza, Gelman, and Auerbach demonstrated this with unusual precision for American presidential voting, showing that the political events experienced roughly between ages fourteen and twenty-four leave a lifelong imprint on a cohort’s partisan tendencies (Ghitza, Gelman, and Auerbach 2023). On this account, the apparent conservatism of the old is not a drift; it is the fossilised radicalism or conservatism of their own formative decades, carried forward while the world moved on around them. Karl Mannheim had intuited exactly this a century ago in his essay on generations (Mannheim 1928): the decisive political fact about a person is often the historical moment at which they came of age.
But it would be a different kind of error — and a violation of this series’ rule against suppressing contrary evidence — to claim that there is no genuine life-cycle component. There is, and it is real if secondary. Studies that separate age, period, and cohort effects using panel data do find some true ageing effect toward conservative voting alongside the larger cohort effect; Tilley and Evans, working with British data, identified both a generational and a life-cycle contribution to rising Conservative support with age (Tilley and Evans 2014). The honest synthesis is therefore this: the dominant force is persistence of formative views (a cohort effect), but there is a genuine, smaller, life-cycle drift toward the status quo, and the interesting question is what drives that residual drift.
The most defensible answer is not psychological but economic, and it is the bridge to the argument of the earlier essays. As people age they move, on average, from being net debtors and non-owners to being net asset holders — owners of homes and accumulated pensions. That transformation changes their material interests in a systematic direction: toward the protection of asset values, toward low inflation and low taxation of wealth, toward the stability of the existing distribution from which they now benefit, and away from the redistributive and disruptive politics that appeal to those who own little. A secondary strand is risk: there is evidence that risk-aversion rises with age, which favours the known over the untried and the stable over the transformative. Life-cycle conservatism, to the extent it is real, is thus substantially the politics of accumulated property and increasing caution, not a mysterious rightward tropism. This matters enormously for what follows, because it means an ageing society is one in which a growing share of the decisive electorate has a direct material stake in exactly the asset-protecting, redistribution-resisting politics that the wealth-defence industry of the plutocracy argument also serves. The old and the rich are not the same people, but on the central question of protecting accumulated wealth their interests increasingly rhyme.
The economic self-interest of age
Follow that residual, asset-driven conservatism into public finance and the political economy of an ageing society comes into focus, and with it a form of inequality the standard rich-poor framing misses entirely: the growing disparity between generations.
An ageing electorate whose most reliable voters are asset-holders and benefit-recipients exerts a predictable gravitational pull on the state’s priorities. Spending tilts toward the programmes that serve the old — pensions and health and long-term care — and away from the investments that serve the young and the future, above all education and the public capital that raises long-run productivity. James Poterba’s analysis of American public education found precisely this pattern: jurisdictions with larger elderly populations spent less per pupil, a finding with an uncomfortable further dimension where the elderly and the school-age populations differed in ethnicity (Poterba 1997). The mechanism is not malice; it is the ordinary politics of a self-interested and reliably voting constituency. But the cumulative effect is a state that transfers resources from young to old on a scale that the language of “the welfare state” obscures, and that under-invests in the very things — education, research, infrastructure, family formation — on which the young and the society’s future capacity depend.
The intergenerational transfer runs through private wealth as well as public budgets, and here Britain is the sharpest case. Because the dominant form of both wealth and old-age security in the UK is housing, and because housing wealth is held overwhelmingly by cohorts who bought property decades ago, the appreciation of house prices has functioned as an enormous, untaxed transfer to the old at the expense of the young, who face housing costs that consume a share of income their parents never contemplated and who are, in growing numbers, locked out of ownership altogether. The generational accounting tradition associated with Auerbach, Gokhale, and Kotlikoff was built precisely to make these hidden intergenerational balances visible — to show that the fiscal and asset structures of ageing societies embody large, implicit transfers across cohorts that conventional accounting does not display (Auerbach, Gokhale, and Kotlikoff 1991). Thomas Piketty’s account of a world in which the return on capital exceeds the growth rate, so that inherited and accumulated wealth outpaces earned income, describes the same phenomenon from another angle (Piketty 2014): in a low-growth, ageing society, the advantage shifts decisively from those who must earn to those who already own, and the old own.
The result is a disparity that the plutocracy essays touched but did not fully name. Alongside the concentration of wealth at the top there is a concentration of wealth among the old, and the two overlap and reinforce each other. The young are disadvantaged not only if they are poor but simply by being young, because they are on the wrong side of both an asset gap and a fiscal structure oriented toward those who came before them. An ageing society is, other things equal, one in which the accident of birth-cohort becomes an increasingly powerful determinant of life chances — which is to say, one in which the open-access ideal of the earlier essays, where opportunity does not depend on the circumstances of one’s origin, quietly erodes along a generational rather than a purely economic seam.
Immigration as the demographic release valve
Against this tide, immigration is the single most powerful lever a society possesses, and the sharpest distinction in the ageing world is between the societies that pull it and the societies that refuse to.
The mechanics are simple. Immigrants are, on average, younger than the native population and arrive at working age, which means they enter the labour force and the tax base immediately without the state having borne the cost of raising and educating them; they often, at least in the first generation, have higher fertility; and they therefore slow both the ageing of the workforce and the deterioration of the dependency ratio on which public finances turn. A rich country facing sub-replacement fertility can maintain a growing or stable working-age population, and thus fund the pensions and health care of its own swelling elderly cohort, essentially only through immigration. This is not a marginal effect. It is the difference between the American trajectory and the Japanese one.
The distinction between the two responses is the key comparative fact. The United States, the United Kingdom, Canada, and — since the mid-2010s and after acute controversy — Germany have used immigration, whether by design or by drift, to offset demographic decline, and it is why the American age structure, though ageing, remains materially younger than the German or Italian and far younger than the Japanese. Against them stand the societies that have refused the lever: Japan and South Korea, which have kept immigration low as a matter of settled preference and now age with extraordinary speed and no demographic relief; much of Eastern and Southern Europe, which combines low fertility with net emigration of its own young; and, most consequentially, China. China is not, and has never structured itself to be, a country of immigration. It faces the fastest large-scale ageing in history, at middle income, with a shrinking population and a workforce that has already begun to contract, and it has no equivalent of the American or German inflow to cushion the blow. The demographic trap that Japan entered as a wealthy society, China is entering as a much poorer one, and without the option — cultural, political, and practical — of importing the young workers it will need. This is the deepest structural difference between the Chinese predicament and the Western one, and it is why China’s ageing is not merely a version of the rich world’s but a distinct and harder problem.
Two honest qualifications are required. First, immigration slows ageing but does not stop it: immigrants themselves age, and the fertility of the second generation tends to converge toward the native norm, so immigration is a way of buying time and sustaining the working-age population at the margin, not a permanent escape from the underlying arithmetic. To hold a dependency ratio constant by immigration alone would require inflows that rise without limit, which is neither plausible nor intended. Second, the demographic benefit is real but it is not the whole of the story, because immigration does not enter a society only as an adjustment to its age structure. It enters as people, and its effects on inequality, on solidarity, and on the character of the population are the subject of the next two sections — and they are the reason the lever is so politically fraught.
Immigration, disparity, and the welfare state
Immigration relieves the economics of ageing, but the same essays’ concern with disparity requires asking what it does to the distribution of income and to the solidarity that funds redistribution — and here the evidence is genuinely mixed and must be presented as such rather than resolved by assertion.
Take the labour market first, because it is the most contested question in the field. The pessimistic view, associated above all with George Borjas, holds that immigration increases the supply of labour and therefore, by ordinary price theory, depresses the wages of the native workers who compete most directly with immigrants — typically the low-skilled — even if it raises aggregate output and benefits capital and consumers (Borjas 2003). The optimistic view, associated with David Card, finds in the empirical record surprisingly little wage effect even from large and sudden inflows, because immigrants also create demand, complement rather than substitute for many native workers, and are absorbed by capital adjustment and shifts in the industrial mix (Card 1990; Card 2005). This dispute is not settled, and honesty requires saying so: the wage effect of immigration on competing natives is real in theory, modest and hard to detect in much of the evidence, and dependent on the skill composition of the inflow and the flexibility of the receiving economy. What can be said is that immigration tends to raise dispersion — it lifts aggregate output and the returns to capital and to complementary skilled labour, while its effects at the bottom are at best neutral and at worst mildly negative — so that even on the optimistic reading it more plausibly widens than narrows the distribution of market income.
The deeper channel to disparity, however, runs not through wages but through the politics of redistribution, and here the evidence is stronger and more troubling. Alberto Alesina and Edward Glaeser argued, in one of the most influential comparative studies in political economy, that the smaller size of the American welfare state relative to the European is explained substantially by the greater racial and ethnic heterogeneity of the United States: solidarity, and the willingness to fund redistribution to strangers, is easier to sustain within a population that sees itself as ethnically homogeneous and harder where recipients are perceived as belonging to a different group (Alesina and Glaeser 2004). The same authors’ work with Baqir and Easterly found, within the United States, that ethnically fragmented jurisdictions provide fewer public goods (Alesina, Baqir, and Easterly 1999). The mechanism is uncomfortable but well documented: ethnic heterogeneity tends to erode the sense of common membership on which generous redistribution depends. Alesina, Miano, and Stantcheva sharpened the point by showing that natives systematically overestimate the number and the poverty of immigrants, and that priming these misperceptions reduces support for redistribution (Alesina, Miano, and Stantcheva 2023).
The implication for an ageing society is a genuine bind, and it is the crux of why immigration and disparity are linked. The lever that relieves the economics of ageing — importing young workers to sustain the workforce and the tax base — is the same lever that, by increasing heterogeneity, tends to weaken the solidarity that funds the redistributive state, and thus to widen inequality on the very dimension the welfare state exists to compress. And it adds a further seam of division that demographers have made central to the American case: the old are disproportionately of the long-established ethnic majority while the young and the immigrant-descended are disproportionately of newer and more diverse groups, so that the intergenerational contract — the young taxed to support the old — increasingly runs across an ethnic line as well as an age line (Frey 2015). A society in which the taxpayers funding the pensioners are visibly a different group from the pensioners is a society in which the solidarity underpinning the whole edifice is under strain from two directions at once. Immigration does not simply buffer ageing. It trades a demographic problem for a distributional and political one, and the trade is real.
Immigration and the character of a population
Beyond income and public finance lies the question the earlier essays raised under the heading of dynamism and social capital: what immigration and the ageing it offsets do to the character of a population — its trust, its cohesion, its capacity for collective action, and its politics. This is the most delicate terrain in the essay, and the one where careful, two-sided treatment matters most.
The uncomfortable empirical starting point is Robert Putnam’s study of American communities, which found that, in the short to medium run, greater ethnic diversity was associated with lower social trust — not merely lower trust across ethnic lines but lower trust generally, including within groups, so that people in more diverse communities tended to “hunker down,” withdrawing from civic life and trusting their neighbours less (Putnam 2007). This finding was and remains genuinely disquieting to those, Putnam included, who value both diversity and social solidarity, and it must be stated plainly rather than explained away. But it must be stated with its full context, which points in a more hopeful direction and is part of Putnam’s own argument. First, Putnam framed the effect as a short-run phenomenon and argued that successful societies, over the longer run, construct broader identities that dissolve the initial withdrawal — that diversity’s costs are transitional and its benefits durable if integration succeeds. Second, the finding has been seriously challenged: Abascal and Baldassarri, re-analysing the data, argued that much of the apparent diversity-trust relationship is driven by economic disadvantage and by the racial composition of communities rather than by diversity as such, so that the pure effect of heterogeneity, net of poverty and of majority-minority dynamics, is far weaker than the headline suggests (Abascal and Baldassarri 2015). The honest reading is that heterogeneity can strain trust and cohesion in the short run, especially where it coincides with disadvantage, but that the effect is contested in magnitude and plausibly transitional where integration is achieved.
Overlay this on an ageing native population and the political effect the earlier essays described comes into view. Pippa Norris and Ronald Inglehart’s account of the authoritarian-populist backlash locates its core in exactly this interaction: older cohorts, formed in more homogeneous and traditional decades, reacting against rapid cultural change and rising diversity, and mobilising around a politics of restoration and control (Norris and Inglehart 2019). The demographic reading of the populist surge across the Western democracies is thus not that immigration alone caused it, nor that ageing alone did, but that the conjunction did — an ageing, formatively-conservative native electorate encountering the visible diversity produced by the immigration used to offset its own decline, and responding with the backlash politics that, as the second essay in this series argued, has expressed itself partly in the legal narrowing of protest, franchise, and dissent. Immigration, taken to relieve the economics of ageing, feeds a politics that erodes the very liberal-democratic mechanisms the earlier essays were concerned to defend.
Yet the same terrain contains the strongest single counter to the pessimism of this whole essay, and it must be given its due. Immigration does not only strain cohesion; it restores dynamism. Immigrants are, across a wide body of evidence, disproportionately entrepreneurial, founding new firms and filing patents at rates above the native-born, and skilled immigration in particular has been a central engine of innovation in the economies that attract it (Kerr 2018). The lever that relieves ageing and strains solidarity is also the lever that counteracts the loss of economic dynamism that ageing otherwise produces. This is why the distinction between the societies that accept immigration and those that refuse it is so consequential in both directions: the acceptors buy demographic time and entrepreneurial vitality at the price of heightened distributional and cultural conflict, while the refusers — Japan, Korea, and above all China — avoid the conflict at the price of ageing faster, stagnating sooner, and forgoing the one proven source of renewed dynamism available to them.
Less dynamism: the growth and innovation channel
That brings the argument to dynamism directly, and to a debate that is live and unresolved in economics, so that the claim “ageing reduces dynamism” must be stated with its genuine qualifications rather than as a settled result.
The oldest version of the idea is Alvin Hansen’s, who in 1938 coined “secular stagnation” precisely to argue that declining population growth would sap investment demand and hold economies below their potential, because a shrinking and ageing population needs less new housing, less new capital, and less of everything that population growth calls forth (Hansen 1939). Lawrence Summers revived the thesis in the 2010s to explain the persistently weak demand and low interest rates of the post-crisis rich world, placing demographic slowdown among its central causes (Summers 2015). The intuition is that a society which is not growing in numbers, and whose members are increasingly old, generates weaker demand for investment and channels more of its resources into consumption by the retired, with slower growth as the result.
The direct empirical evidence on ageing and growth is genuinely divided, and both sides must be reported. On one side, Maestas, Mullen, and Powell, exploiting variation across American states, estimated that population ageing substantially reduced economic growth, with the largest part of the effect operating not through the shrinking labour force alone but through slower productivity growth in an older workforce (Maestas, Mullen, and Powell 2023). On the other side — and this is the contrary evidence the standard of this series forbids omitting — Daron Acemoglu and Pascual Restrepo, examining variation across countries, found no negative relationship between ageing and growth in GDP per head, and if anything a positive one, and argued that the societies ageing fastest have offset the loss of labour by adopting automation more aggressively, so that robots and machines substitute for the scarce young workers (Acemoglu and Restrepo 2017; Acemoglu and Restrepo 2022). These findings are not easily reconciled, and it would be dishonest to pretend otherwise: the state-level American evidence suggests ageing bites hard through productivity, while the cross-country evidence suggests that automation can neutralise the labour-supply effect. The most that can be said with confidence is that ageing exerts a downward pressure on growth through labour supply and, on some evidence, productivity, but that this pressure can be partly or wholly offset by capital deepening and automation, so that the net effect on aggregate output is contested.
Where the evidence is clearer is on the narrower question of business dynamism — the churn of firm entry and exit and the reallocation of resources to new and growing enterprises that drives innovation and productivity in the long run. American business dynamism has been declining for decades, with fewer new firms formed and a rising share of activity in established incumbents (Decker et al. 2014), and the demographic connection has now been drawn explicitly. Karahan, Pugsley, and Şahin argued that the slowdown in labour-force growth caused by demographic change is a principal cause of the declining startup rate, because a slower-growing supply of workers reduces the rate at which new firms are formed (Karahan, Pugsley, and Şahin 2024). Liang, Wang, and Lazear identified a complementary mechanism operating on individuals: in an ageing society the senior positions are occupied for longer by a large older cohort, clogging the “career escalator” and denying younger people the experience and the openings that entrepreneurship requires, so that ageing depresses entrepreneurship not only in the old but in the young (Liang, Wang, and Lazear 2018).
Two findings complicate the popular picture and are owed in the interest of accuracy. First, the association of innovation with youth is overstated: Benjamin Jones showed that great inventors do their major work in middle age, typically in their late thirties or early forties, and that the age of peak invention has been rising (Jones 2010); Azoulay, Jones, Kim, and Miranda found that the founders of the fastest-growing new firms average around forty-five, not the mythical twenty-something (Azoulay et al. 2020). Dynamism is not the exclusive property of the young, and a society of the middle-aged is not thereby innovation-dead. Second, as the previous section noted, immigration can restore firm formation and innovation, so an ageing society that accepts immigrants can offset part of the dynamism loss that an ageing closed society cannot. The defensible conclusion, then, is neither the alarmist “ageing kills growth” nor the complacent “automation makes ageing costless,” but something more precise: ageing reliably depresses business dynamism through labour-supply and career-structure channels, exerts a contested downward pressure on aggregate growth that automation may offset, and does so most severely in the closed societies that forgo the immigration and, sometimes, the automation that could compensate. China, ageing fastest, at middle income, with a shrinking workforce and without the immigration lever, is the society most exposed to the pessimistic reading and least equipped to escape it.
The synthesis: how the pieces lock together
Assemble the mechanisms and a coherent structure appears, one that deepens and extends the plutocracy argument of the earlier essays by supplying its demographic foundation.
An ageing society is one in which the demographic fuel of upheaval is spent: the youth bulge that supplies revolutionary energy has vanished, the biographically available young who staff high-risk collective action are a shrinking cohort, and the decisive electorate is drawn increasingly from the old. It is a society whose politics tilts, through the residual life-cycle conservatism of asset accumulation and risk-aversion and through the sheer voting weight of the old, toward the protection of accumulated wealth, the stability of the existing distribution, and the resistance of redistribution and disruption — which is to say, toward exactly the wealth-defence politics that the plutocracy essays identified, now reinforced by the material interests of a large and reliably voting elderly constituency rather than by the concentrated rich alone. It is a society that transfers resources from young to old through both public budgets and private asset appreciation, producing a disparity along the generational seam that compounds the disparity along the wealth seam, so that the accident of birth-cohort joins the accident of birth-class as a determinant of life chances. And it is a society whose economic dynamism is under downward pressure — fewer new firms, a clogged career escalator, contested but real drag on productivity and growth — which entrenches incumbents, dampens the creative destruction that redistributes economic power, and thereby locks in the existing distribution rather than churning it.
Each of these effects points the same way: toward the entrenchment of the settled order and the enervation of the forces that might disturb it. The plutocracy essays described a ratchet in which concentrated wealth shapes rules that weaken countervailing power that permits further concentration. Demography turns that ratchet more easily. An old, asset-holding, risk-averse, disruption-resistant electorate is a weaker source of countervailing power than a young one; a low-dynamism economy generates fewer of the new fortunes and new firms that might challenge incumbents; a fiscal structure oriented to the old under-invests in the education and opportunity that might equip the young to contest the order. The demographic engine does not create the plutocratic tendency, but it removes friction from it at every point, and it does so through the quiet, cumulative arithmetic of age rather than through any decision anyone takes.
Immigration is the great modifier, and its effect is genuinely double. Where it is accepted, it relieves the economic pressure — sustaining the workforce, the tax base, and, through immigrant entrepreneurship, some of the lost dynamism — but it inflames the politics, straining the solidarity that funds redistribution, adding an ethnic seam to the generational one, and feeding the backlash politics that, as the second essay argued, has narrowed the liberal-democratic mechanisms of dissent. Where it is refused, the politics of heterogeneity is avoided but the economic reckoning is faced in full: faster ageing, sooner stagnation, and no external source of renewal. The Western democracies have chosen, mostly by drift, the first path and now live with its distributional and cultural conflicts; Japan and Korea have chosen the second and age in relative social peace toward economic stasis; China, ageing fastest and poorest, has the second path forced upon it and faces the harshest version of the trap. The common demographic process yields divergent national predicaments, but the underlying logic — that age entrenches, that disparity deepens along new seams, that dynamism fades — runs through all of them.
The honest counter-case
The standard of this series requires that the argument be tested against its strongest opposition, and there is genuine, serious opposition that a candid account must present at full strength rather than in caricature.
The most powerful challenge to the disparity thesis comes from Charles Goodhart and Manoj Pradhan, who argue that the coming decades will reverse, not extend, the inequality of the recent past (Goodhart and Pradhan 2020). Their reasoning is that the great disinflation and the great compression of labour’s bargaining power since the 1980s were driven above all by a vast positive labour-supply shock — the entry of China and Eastern Europe into the global economy and the arrival of the baby-boomers into the workforce — and that this shock is now reversing. As workforces shrink and age, labour becomes scarce, and scarce labour commands higher wages and greater bargaining power, so that the coming demographic era should see labour’s share rise, inequality between capital and labour narrow, and inflation return. If they are right, ageing does not deepen disparity but compresses it, and the causal arrow of this essay’s disparity section points the wrong way. This is a formidable argument from serious economists, and it cannot be dismissed; the most that can be said in reply is that it concerns primarily the capital-labour and wage distribution and has less to say about the intergenerational and asset-driven disparities that this essay has emphasised, that its predicted labour-scarcity effect can be blunted precisely by the automation Acemoglu and Restrepo document and by immigration, and that its central prediction of returning inflation remains an open empirical question. It weakens the disparity claim in one important dimension; it does not obviously overturn the intergenerational and asset dimensions. But it must be weighed, and readers should weigh it.
The dynamism thesis faces the Acemoglu-Restrepo challenge already presented: the cross-country evidence that ageing has not reduced growth per head and that automation offsets the labour-supply loss. If that evidence is the better guide than the state-level American findings of Maestas and colleagues, then the “less dynamism” claim is weaker than the essay allows, at least at the level of aggregate output, and the business-dynamism decline, real as it is, may not translate into the growth stagnation the pessimistic reading fears.
The conservatism thesis, as the relevant section already conceded, is largely a cohort effect rather than a life-cycle one, and this cuts against any simple demographic pessimism in a specific and important way: if the rightward tendency of the old is mostly the persistence of their formative views rather than a product of ageing as such, then it is not a permanent property of ageing societies but a feature of which cohorts are currently old. As the generations formed in more liberal and diverse decades age into the electorate, the political character of the old will change with them, and the association between age and conservatism observed today may weaken or even reverse. The demographic entrenchment of the settled order is therefore less mechanical and more historically contingent than a determinist reading would have it, and that contingency is a genuine ground for doubting the gloomier conclusions.
And the immigration-disparity link, as its section stressed, rests partly on the contested labour-market evidence, where Card’s findings of small wage effects sit against Borjas’s, and on the diversity-trust evidence, where Abascal and Baldassarri’s re-analysis substantially weakens Putnam’s headline. None of these counters is decisive, but together they establish that the pessimistic synthesis is a well-supported tendency rather than an iron law, and that responsible readers should hold it as the more likely reading of contested evidence, not as demonstrated fact.
Conclusion: the quiet arithmetic
The tide described in this essay moves too slowly to be news and too surely to be ignored. Fertility across the rich world and much of the developing world has fallen below the level that replaces a population; the societies built on that fertility are ageing and, increasingly, shrinking; and the effects of that ageing, worked through participation, self-interest, public finance, and enterprise, compose a recognisable pattern. Ageing societies are structurally unlikely to produce revolutions, because the demographic fuel of upheaval — the young, biographically available, low-opportunity-cost cohort — is spent. Their protest, though it can be large, matters less, because the decisive electorate ages toward stability and the institutions that converted mobilisation into power have decayed. Their politics tilts toward the protection of accumulated wealth, less through a mysterious conservatism of age than through the material interests of a growing class of elderly asset-holders whose stake in the existing distribution rhymes with that of the concentrated rich. Their inequality deepens along a generational seam that compounds the seam of wealth, so that the accident of when one was born joins the accident of to whom. And their economic dynamism faces a real downward pressure — fewer new firms, a clogged path for the young, contested but genuine drag on productivity — that entrenches incumbents and slows the churn by which economic power is redistributed.
Immigration is the one lever that changes this arithmetic, and it changes it in both directions at once, relieving the economics of ageing while inflaming its politics — sustaining the workforce and restoring some dynamism at the price of strained solidarity, an ethnic seam laid over the generational one, and a backlash that erodes the liberal-democratic mechanisms the earlier essays sought to defend. The societies that accept the lever and those that refuse it face different predicaments, but the same underlying logic: China ageing fastest and poorest with the lever denied it, Japan and Korea ageing in social peace toward stasis, the West ageing with immigration’s relief and immigration’s conflicts both. The common process yields divergent national fates and a shared direction.
This is not a prophecy of collapse, and the counter-case is real: labour scarcity may yet compress inequality where automation and immigration do not blunt it, automation may yet offset the growth drag, and the conservatism of the old, being mostly a matter of which cohorts are old, will change as the cohorts do. The honest verdict is a tendency, not a law. But the tendency is coherent and it points one way — toward a world that is older, calmer in the specific sense of being harder to move, more unequal across generations as well as classes, and less dynamic; a world in which the settled order is entrenched less by the design of any elite than by the quiet arithmetic of age. The plutocracy essays warned of a door closing on political equality. This essay adds that demography is one of the hands easing it shut, slowly, and without anyone deciding that it should.
Works Cited
Abascal, M., & Baldassarri, D. (2015). Love Thy Neighbor? Ethnoracial Diversity and Trust Reexamined. American Journal of Sociology, 121(3), 722–782.
Acemoglu, D., & Restrepo, P. (2017). Secular Stagnation? The Effect of Aging on Economic Growth in the Age of Automation. American Economic Review, 107(5), 174–179.
Acemoglu, D., & Restrepo, P. (2022). Demographics and Automation. Review of Economic Studies, 89(1), 1–44.
Alesina, A., Baqir, R., & Easterly, W. (1999). Public Goods and Ethnic Divisions. Quarterly Journal of Economics, 114(4), 1243–1284.
Alesina, A., & Glaeser, E. L. (2004). Fighting Poverty in the US and Europe: A World of Difference. Oxford University Press.
Alesina, A., Miano, A., & Stantcheva, S. (2023). Immigration and Redistribution. Review of Economic Studies, 90(1), 1–39.
Auerbach, A. J., Gokhale, J., & Kotlikoff, L. J. (1991). Generational Accounts: A Meaningful Alternative to Deficit Accounting. Tax Policy and the Economy, 5, 55–110.
Azoulay, P., Jones, B. F., Kim, J. D., & Miranda, J. (2020). Age and High-Growth Entrepreneurship. American Economic Review: Insights, 2(1), 65–82.
Borjas, G. J. (2003). The Labor Demand Curve Is Downward Sloping: Reexamining the Impact of Immigration on the Labor Market. Quarterly Journal of Economics, 118(4), 1335–1374.
Card, D. (1990). The Impact of the Mariel Boatlift on the Miami Labor Market. Industrial and Labor Relations Review, 43(2), 245–257.
Card, D. (2005). Is the New Immigration Really So Bad? The Economic Journal, 115(507), F300–F323.
Decker, R., Haltiwanger, J., Jarmin, R., & Miranda, J. (2014). The Role of Entrepreneurship in US Job Creation and Economic Dynamism. Journal of Economic Perspectives, 28(3), 3–24.
Eberstadt, N. (2024). The Age of Depopulation: Surviving a World Gone Grey. Foreign Affairs, 103(6).
Frey, W. H. (2015). Diversity Explosion: How New Racial Demographics Are Remaking America. Brookings Institution Press.
Ghitza, Y., Gelman, A., & Auerbach, J. (2023). The Great Society, Reagan’s Revolution, and Generations of Presidential Voting. American Journal of Political Science, 67(3), 520–537.
Goldstone, J. A. (1991). Revolution and Rebellion in the Early Modern World. University of California Press.
Goodhart, C., & Pradhan, M. (2020). The Great Demographic Reversal: Ageing Societies, Waning Inequality, and an Inflation Revival. Palgrave Macmillan.
Greenhalgh, S. (2008). Just One Child: Science and Policy in Deng’s China. University of California Press.
Hansen, A. H. (1939). Economic Progress and Declining Population Growth. American Economic Review, 29(1), 1–15.
Heinsohn, G. (2003). Söhne und Weltmacht: Terror im Aufstieg und Fall der Nationen. Orell Füssli.
Jones, B. F. (2010). Age and Great Invention. Review of Economics and Statistics, 92(1), 1–14.
Karahan, F., Pugsley, B., & Şahin, A. (2024). Demographic Origins of the Startup Deficit. American Economic Review, 114(7), 1986–2023.
Kerr, W. R. (2018). The Gift of Global Talent: How Migration Shapes Business, Economy & Society. Stanford University Press.
Krosnick, J. A., & Alwin, D. F. (1989). Aging and Susceptibility to Attitude Change. Journal of Personality and Social Psychology, 57(3), 416–425.
Liang, J., Wang, H., & Lazear, E. P. (2018). Demographics and Entrepreneurship. Journal of Political Economy, 126(S1), S140–S196.
Maestas, N., Mullen, K. J., & Powell, D. (2023). The Effect of Population Aging on Economic Growth, the Labor Force, and Productivity. American Economic Journal: Macroeconomics, 15(2), 306–332.
Mannheim, K. (1928). Das Problem der Generationen. Kölner Vierteljahrshefte für Soziologie, 7. (English: The Problem of Generations, in Essays on the Sociology of Knowledge, 1952.)
McAdam, D. (1988). Freedom Summer. Oxford University Press.
Norris, P., & Inglehart, R. (2019). Cultural Backlash: Trump, Brexit, and Authoritarian Populism. Cambridge University Press.
Piketty, T. (2014). Capital in the Twenty-First Century (A. Goldhammer, Trans.). Harvard University Press.
Poterba, J. M. (1997). Demographic Structure and the Political Economy of Public Education. Journal of Policy Analysis and Management, 16(1), 48–66.
Putnam, R. D. (2007). E Pluribus Unum: Diversity and Community in the Twenty-first Century. Scandinavian Political Studies, 30(2), 137–174.
Sears, D. O., & Funk, C. L. (1999). Evidence of the Long-Term Persistence of Adults’ Political Predispositions. Journal of Politics, 61(1), 1–28.
Summers, L. H. (2015). Demand Side Secular Stagnation. American Economic Review, 105(5), 60–65.
Tilley, J., & Evans, G. (2014). Ageing and Generational Effects on Vote Choice: Combining Cross-Sectional and Panel Data to Estimate APC Effects. Electoral Studies, 33, 19–27.
Urdal, H. (2006). A Clash of Generations? Youth Bulges and Political Violence. International Studies Quarterly, 50(3), 607–629.
Data sources: World Bank, World Development Indicators — total fertility rate (SP.DYN.TFRT.IN), population ages 65+ (SP.POP.65UP.TO.ZS), age dependency ratio, old (SP.POP.DPND.OL), and total population (SP.POP.TOTL), drawn from the UN Population Division’s World Population Prospects, for China, the United States, the United Kingdom, Germany, Italy, Japan, South Korea, India, and Nigeria, 1960–2025.