The Weight of Years
Why ageing societies rebel less, calcify more, and slide toward entrenched inequality — across the West, and most sharply in China.
The current beneath the politics
Two earlier essays in this sequence argued that the United States and Britain are not plutocracies but sit in the antechamber of one: the money has concentrated to levels not seen in a century, the legal cost of collective resistance has risen, and the organised countervailing power of ordinary people has hollowed out. That argument left a question unanswered. Why now? What is the deep current carrying two mature democracies toward a condition their institutions were built to prevent?
Part of the answer is demographic, and the purpose of this essay is to trace it — carefully, because demography is the field where confident storytelling most often outruns the evidence. The claim I will defend is bounded. It is not that demography is destiny; it is that demography is a current, something that changes the odds of political and economic outcomes without determining them. An ageing population, I will argue, exhibits a lower propensity for disruptive collective action for reasons that are structural rather than mysterious; it shifts the political centre of gravity toward the defence of the status quo through a mechanism that does not depend on anyone’s mind changing; and it is associated — through several channels, some better evidenced than others — with lower economic dynamism and with the entrenchment of existing wealth. Immigration is the great modulator of all of this: it offsets the economics of ageing while igniting the cultural politics that populist movements ride, and its effect on inequality runs through channels that are frequently misidentified.
I want to be exact about the epistemic status of each link, because several of them are genuinely contested in the peer-reviewed literature, and the temptation in an essay like this is to present one side and suppress the other. I will not do that. Where the evidence is mixed — on whether ageing reduces growth, on whether immigration lowers wages, on whether ageing raises or lowers within-country inequality — I will state both sides and name the studies on each, and I will let the disagreement stand rather than paper over it. The synthesis I am after does not require every link to be certain. It requires only that the links, taken together and weighted by what the evidence actually supports, describe a world that is becoming less rebellious, less dynamic, and more unequal — and that the sharpest version of the whole process is unfolding not in the West at all, but in China.
The measurement trap: period, cohort, and life-cycle
Before any of the substantive claims, a methodological warning, because ignoring it is the most common way demographic arguments become fraudulent. When we observe that older people today hold more conservative views, or vote more, or protest less, we are observing a correlation between age and behaviour at a single point in time, and that correlation can arise from three completely different processes that a cross-section cannot distinguish (Ryder 1965; Mannheim 1928/1952).
The first is a life-cycle effect: people change as they age, so that the same individual becomes more conservative, or more risk-averse, or less politically active, simply by growing older. The second is a cohort effect: different generations carry different attitudes formed in their youth and retain them through life, so that what looks like the effect of age is actually the persistence of the formative experiences of people born at a particular time. The third is a period effect: everyone alive changes together in response to contemporary events, and the appearance of an age pattern is an artefact of when the measurement was taken. These three are not merely academically distinct. They imply opposite futures. If older people are conservative because ageing makes people conservative, then an ageing society becomes more conservative mechanically. If older people are conservative because a particular conservative cohort happens to be old right now, then the politics of the elderly will change entirely as more liberal cohorts age into them, and the inference collapses.
The cross-sectional data alone cannot separate these — this is the age-period-cohort identification problem, and it has no purely statistical solution, because age, period, and cohort are exactly collinear (period minus age equals cohort). Any claim that “older people are X, therefore an ageing society will be more X” that does not confront this problem is resting on a hidden and untested assumption, and by the standard I am applying it is a defect. What can be said requires panel data that follow the same individuals over time, and the evidence there is more equivocal than the popular story admits. The most robust finding is not that ageing reliably shifts attitudes but close to the opposite: that political predispositions are formed early — the “impressionable years” of roughly late adolescence to the mid-twenties — and then become substantially more resistant to change, though not immovable, across the rest of life (Krosnick and Alwin 1989; Sears and Funk 1999; Ghitza, Gelman, and Auerbach 2023). This finding cuts against the naive life-cycle story and toward the cohort story, and it means that the strongest version of “ageing makes people conservative” is not well supported. I will return to what is supported — a stakes-based mechanism that does not require attitudes to change at all — after establishing the scale of the demographic shift itself.
How old the world is getting
The demographic transition is not a forecast; the part that matters for this argument has already happened, and it is documented in the World Bank’s compilation of national and UN population statistics. Fertility has fallen below the replacement rate of roughly 2.1 births per woman across essentially the entire developed world and much of the developing world. As of 2024, the total fertility rate stood at about 1.63 in the United States, 1.55 in the United Kingdom, 1.61 in France, 1.43 in Sweden, 1.36 in Germany, 1.18 in Italy, and 1.15 in Japan (World Bank, indicator SP.DYN.TFRT.IN). In East Asia the collapse is more extreme still: South Korea recorded a total fertility rate of about 0.75, the lowest of any sizeable nation in recorded history, and China about 1.01 — down from 4.45 as recently as 1960. Even India, long the demographic counterweight, had fallen to about 1.96 by 2024, below replacement for the first time.
Sustained sub-replacement fertility, working through time, produces population ageing with the reliability of arithmetic. The share of the population aged 65 and over now stands at roughly 30% in Japan, 25% in Italy, 24% in Germany, 22–23% in France, 21% in Sweden, 20% in South Korea, 20% in the United Kingdom, and 18% in the United States (World Bank, SP.POP.65UP.TO.ZS, 2025 values). The old-age dependency ratio — the number of people aged 65 and over for every hundred of working age — tells the same story in the form that matters fiscally and politically: about 51 in Japan, 40 in Italy, 38 in Germany, 37 in France, 34 in Sweden, 31 in the United Kingdom, 29 in South Korea, and 29 in the United States (SP.POP.DPND.OL, 2025). Half a century ago these ratios sat, in most of these countries, in the mid-teens to low twenties. They have roughly doubled, and — because they are driven by fertility decisions made decades ago and by continued gains in longevity — they will keep rising for decades more, whatever fertility does from here.
China is the case to hold in view throughout, because it combines speed, scale, and an absence of the escape valve the West relies on. In 1980 only about 4.4% of Chinese were over 65; by 2025 the figure is about 14.9%, and it is climbing steeply. China’s working-age share of the population peaked around 2010 at roughly 73% and has been falling since. Its total population peaked around 2021–2022 at just over 1.41 billion and has since begun to decline, so that India overtook it as the world’s most populous country around 2022–2023 (World Bank, SP.POP.TOTL). No large society has ever aged this fast, and none has done so while remaining as closed to immigration. That combination is the reason China, rather than Japan or Italy, is the sharpest test of every mechanism this essay describes.
Why the young revolt and the old do not
Begin with the claim that ageing societies are less prone to upheaval, because it is the best evidenced of the lot and it does not depend on the contested psychology of individual attitude change. It depends instead on the composition of the population and on the structure of incentives that composition creates.
The empirical regularity is well established in the conflict literature: societies with large cohorts of young adults — “youth bulges” — face a systematically elevated risk of political violence, unrest, and civil conflict. Henrik Urdal’s cross-national analysis found that the larger the share of young adults in the adult population, the higher the risk of internal armed conflict and political violence, with the association robust across specifications and controls (Urdal 2006). This sits within a broader tradition — Jack Goldstone’s demographic-structural theory of state breakdown — arguing that rapid growth in the number of young people, particularly educated young people whose opportunities do not match their expectations, is a recurring ingredient in revolutions and rebellions from the early modern period onward (Goldstone 1991; Goldstone 2002). The mechanism is not that young people are inherently violent; it is that a large cohort competing for scarce positions, jobs, and status generates precisely the frustrated ambition that political entrepreneurs can mobilise.
Two further mechanisms explain the individual-level side of the same pattern. The first is biographical availability, a concept from the study of social movements: high-risk, high-cost activism is disproportionately undertaken by people who are free of the countervailing commitments — a career, a mortgage, dependents, a pension — that raise the personal cost of disruption. Doug McAdam’s study of the 1964 Freedom Summer volunteers found that participation in high-risk activism was strongly predicted by biographical availability: the young, the unmarried, those without full-time jobs or dependent children were far more likely to put themselves at risk (McAdam 1986). An ageing society is, almost by definition, a society in which the biographically available share of the population is shrinking. The second mechanism is opportunity cost in its economic form. A person with accumulated assets — a house, a pension, savings — bears a larger expected loss from instability, disorder, and the destruction of value that upheaval brings. The old, as a class, are the holders of accumulated assets. They therefore have, on average, more to lose from disruption and less to gain, and this is true regardless of whether their political opinions have shifted a single degree. The frequently invoked idea that individuals also become more psychologically risk-averse with age is more weakly and inconsistently evidenced, and I do not lean on it; the compositional and stakes-based mechanisms are sufficient and are far better supported. An old society is a quiet society not chiefly because old individuals have lost their appetite for risk, but because there are fewer of the biographically available young, and because the median citizen now holds a stake in the existing order that disruption would put at risk.
Does age make people conservative?
Now the harder and more contested claim, which I will not overstate. In what sense, if any, does an ageing population become politically conservative?
The honest answer begins with the measurement trap of the second section. The cross-sectional fact that older voters lean rightward in many contemporary democracies is real but nearly uninterpretable on its own, because it confounds life-cycle, cohort, and period. The panel evidence that can separate them offers only modest support for a genuine life-cycle drift toward conservatism, and a good deal of support for the competing view that attitudes are largely set young and persist (Krosnick and Alwin 1989; Sears and Funk 1999). Studies that attempt to decompose British and American voting into age, period, and cohort components find that both life-cycle and cohort effects operate, with neither wholly dominant, and that the specific mapping of “old equals conservative” is partly a feature of which cohorts happen to be old at a given moment rather than a law of ageing (Tilley and Evans 2014; Ghitza, Gelman, and Auerbach 2023). The generational-imprint research is especially deflating to the simple story: it suggests that the political colour of the elderly decades hence will be set by the formative experiences of people who are young now, not by the mellowing-into-conservatism of whoever is old today.
So the strong psychological claim — that growing older reliably makes people more conservative — is not something the evidence licenses, and asserting it would be exactly the kind of untested distributional claim that this essay’s standard forbids. What survives scrutiny is narrower and more defensible in two forms. First, there is reasonable evidence for a life-cycle increase in small-c conservatism in the dispositional sense — a growing attachment to stability, familiarity, and the existing order — that is distinct from party allegiance and considerably more robust than any claim about left-right position. Second, and more important for this argument, there is the stakes mechanism, which requires no attitude change whatsoever and to which the next section is devoted. The distinction matters because the popular discourse collapses these into a single lazy claim — “people get more right-wing as they age” — that the data do not support in that form. The defensible claim is that an ageing electorate becomes more status-quo-oriented, and that this can happen even if not one voter changes a single opinion, because the composition of the electorate and the distribution of what its members have to protect are themselves changing.
The stakes model: the asset-holder electorate
Here is the mechanism that does the real work, and its virtue is that it sidesteps the entire psychological debate. Suppose, for the sake of argument, that no individual ever changes a political view in their life — that every person dies holding exactly the opinions they formed at twenty-five. An ageing society would still become more oriented toward the defence of accumulated wealth and the existing order, for two reasons that are matters of arithmetic and revealed interest rather than psychology.
The first is turnout. Across essentially every established democracy, and consistently across decades of data, older citizens vote at markedly higher rates than younger ones (a regularity documented in national election studies and, for the United States and United Kingdom respectively, in the Census Bureau’s voting supplements and the British Election Study). An electorate that is ageing is therefore an electorate whose effective median voter — weighted by the probability of actually casting a ballot — ages faster than the population, so that the preferences politicians must court shift toward those of the old even faster than the raw demography would suggest.
The second is what the old, as an economic class, have an interest in. Older citizens are the holders of the society’s accumulated assets — housing above all, and pension claims — and their revealed political priorities reflect that position: the protection of property values, the protection of retirement income, and stability in the arrangements from which they benefit. This is not a moral failing; it is the ordinary economics of a stage of life in which one has stopped accumulating and started drawing down. But its systemic effect is to align the pivotal voter’s interest with the asset-holding interest — which is substantially the same interest that concentrated wealth defends. In Britain this alignment is unusually tight, because the dominant form of both private wealth and old-age security is the owner-occupied house, so that the interests of the asset-rich elderly and the property-owning wealthy point in the same direction: keep house prices high, resist supply that would lower them, and treat the existing distribution as something to be defended. This is the precise point at which demography meets the plutocracy argument of the earlier essays. A political system in which the most reliable voters are also the principal holders of the assets that concentration protects is a system whose democratic pressure runs, on net, with the grain of concentration rather than against it. The demographic engine does not invent the plutocratic tendency. It removes the friction that would otherwise resist it — and it does so whether or not anyone’s opinions have moved an inch.
Immigration as the demographic offset
Immigration enters this story first as the one available remedy for the arithmetic of ageing, and understanding its demographic role is a prerequisite to understanding its political one. A society with sub-replacement fertility has, in principle, only three responses to a rising old-age dependency ratio: raise fertility, which no wealthy country has reliably managed to do at scale despite decades of pro-natal policy; raise the working share by importing working-age people, which is immigration; or accept the rising ratio and its fiscal consequences. The United Nations Population Division’s much-discussed “replacement migration” analysis made the limits of the second option vivid: the migration inflows required to hold old-age dependency ratios constant in the most rapidly ageing countries would be so large — in some scenarios implying that immigrants and their descendants would come to constitute the majority of the population within decades — as to be politically inconceivable (United Nations Population Division 2000). Immigration can cushion ageing; it cannot, at any politically feasible scale, reverse it.
The distinction that matters most for this essay is between the societies that use immigration as a partial offset and those that do not, because that single policy choice sorts the ageing world into two very different trajectories. The United States, the United Kingdom, Germany (especially after 2015), Canada, and Australia have, to varying degrees and with varying deliberateness, admitted enough working-age migrants to slow the rise in their dependency ratios and to sustain labour-force growth that fertility alone would not provide. Japan, South Korea, and above all China have not. Japan’s working-age share has fallen to about 59%, the lowest among the countries considered here, and it has met the resulting labour shortage largely through automation, higher female and elderly labour participation, and, only recently and reluctantly, modest foreign-worker programmes rather than large-scale settlement. China is the extreme: it is not merely low-immigration but effectively closed to permanent immigration as a matter of both policy and national self-conception, so that as its working-age population shrinks it has no external supply to draw on. This is the first of the distinctions the whole argument turns on. Ageing is universal across the developed world; the presence or absence of an immigration offset is what separates societies that will merely grow older from societies that will also shrink — and shrinking, as China is about to discover, is a harder problem than ageing alone.
Immigration and the backlash
Immigration offsets the economics of ageing, but it does so at a political price, and the price is highest precisely in the ageing native populations that most need the offset. This is the central irony of the demographic politics of our moment, and it is well documented.
The most influential synthesis is the “cultural backlash” thesis of Pippa Norris and Ronald Inglehart, who argue that the rise of authoritarian-populist movements across Western democracies is driven less by economic grievance than by a reaction — concentrated among older, less educated, native-born voters — against the cultural changes of recent decades, immigration prominent among them (Norris and Inglehart 2019; Inglehart and Norris 2017). The evidence on what actually drives anti-immigration sentiment reinforces the point and corrects a common error. The careful survey-experimental and observational work of Jens Hainmueller and Daniel Hopkins finds that opposition to immigration is driven far more by sociotropic and cultural concerns — beliefs about the effect of immigration on the nation, its culture, and its public life — than by personal economic self-interest; the classic prediction that native workers oppose immigration because it threatens their own wages receives weak support (Hainmueller and Hopkins 2014). Relatedly, Hainmueller and Hiscox show that both high- and low-skilled natives tend to prefer high-skilled immigrants, a pattern inconsistent with simple labour-market self-interest and more consistent with concerns about fiscal burden and cultural distance (Hainmueller and Hiscox 2010). The upshot is that the backlash is not, in the main, the revolt of workers protecting wages; it is a reaction, strongest among the old, to perceived changes in the character of the nation.
Layer this onto the demographic structure and the interaction becomes clear and somewhat vicious. The societies that most need immigration to cushion ageing are precisely the societies whose ageing electorates react most strongly against it — and, by the stakes-and-turnout mechanism of the previous section, those electorates are also the most pivotal. The result is a politics in which the old, who benefit from the fiscal support that working-age migrants provide to pension and health systems, nonetheless vote against the immigration that provides it, and in which populist movements convert that reaction into durable electoral coalitions. Crucially — and this connects back to the earlier essays — this backlash energy does not straightforwardly challenge concentrated wealth. It is frequently mobilised by, or captured by, political entrepreneurs who direct it toward cultural and national targets while leaving the distribution of wealth untouched or even regressively altered. Anti-immigration populism is compatible with plutocracy; indeed it can serve as its most effective cover, channelling the discontent of the demographically anxious away from the concentration of wealth and toward the immigrant, the foreigner, and the cosmopolitan elite defined in cultural rather than economic terms.
Immigration and disparity: three channels, honestly distinguished
The claim that immigration increases inequality is frequently made and frequently muddled, because it runs through at least three distinct channels that have very different evidential support. Rigour requires separating them, and honesty requires reporting that the best-known channel is the least settled.
The first and most contested is the wage channel: the argument that immigration, by increasing the supply of labour, depresses wages — particularly at the bottom, widening dispersion. This is the subject of one of the most genuinely unresolved disputes in empirical economics, and I will not pretend it is settled in either direction. David Card’s celebrated study of the 1980 Mariel boatlift, in which a sudden influx of Cuban migrants raised Miami’s labour force sharply, found essentially no measurable negative effect on the wages or employment of existing low-skilled workers (Card 1990), a finding consistent with his broader conclusion that immigration’s effect on the wage distribution is modest (Card 2009). George Borjas, working from a theoretical model in which the labour demand curve slopes down and immigrants and natives of similar skill are close substitutes, argued that immigration does depress the wages of competing native workers, and in a direct reanalysis of the Mariel episode reported significant wage declines for low-skilled Miami workers that Card’s analysis had missed (Borjas 2003; Borjas 2017). Giovanni Peri and Vasil Yasenov, reanalysing the same episode with different methods, contested Borjas’s finding and reaffirmed a null or negligible effect (Peri and Yasenov 2019). This dispute turns on choices about sample composition, comparison groups, and time windows that reasonable economists have not resolved, and I state plainly that the wage-and-inequality effect of immigration is not established — anyone who tells you the labour-market evidence clearly shows large negative wage effects, or clearly shows none, is overstating what the literature supports.
The second channel is compositional and is better grounded. When immigration is skewed toward the high-skilled, its direct effect is to add to the upper part of the distribution; when it is skewed toward the low-skilled, it adds to the lower part; and in either case the arithmetic of the receiving country’s inequality shifts accordingly, before any effect on native wages. High-skill-selective systems and low-skill-heavy inflows therefore have opposite first-order effects on measured inequality, which is one reason cross-national comparisons of “immigration and inequality” so often conflict: they are comparing different kinds of immigration.
The third channel is political, and it is the one with perhaps the cleanest logic and strong supporting evidence, though it is rarely the one people have in mind. Ethnic and national heterogeneity tends to erode support for redistribution: more diverse societies sustain smaller welfare states, because solidarity and the willingness to fund transfers to strangers weaken when the strangers are perceived as different. Alberto Alesina and Edward Glaeser advanced this as a central explanation for why the United States redistributes so much less than Europe (Alesina and Glaeser 2004), and Alesina, Miano, and Stantcheva show experimentally that people systematically overestimate the number and dependency of immigrants and that priming these misperceptions reduces support for redistribution (Alesina, Miano, and Stantcheva 2023). If immigration reduces the political coalition for redistribution, it can raise inequality through the tax-and-transfer system even if its direct labour-market effect on wages is nil. This is the channel most worth attending to, and it is distinct from — and better evidenced than — the wage-competition story that dominates public argument. The honest summary is therefore threefold: the wage channel is real in theory but empirically unresolved; the compositional channel is straightforward but depends entirely on the skill mix; and the political channel, whereby diversity weakens the redistributive coalition, is the most robustly supported and the least discussed.
Ageing and dynamism: the contested case
The claim that ageing societies are less economically dynamic is intuitive, widely asserted, and — this must be said clearly — only partially supported, with at least one major peer-reviewed finding pointing the other way. I will lay out the mechanisms, the supporting evidence, and the contradicting evidence, and I will not resolve by assertion what the literature has not resolved by demonstration.
Several mechanisms plausibly link ageing to reduced dynamism. Entrepreneurship and firm creation depend on the flow of new entrants, and a slower-growing or shrinking working-age population mechanically reduces that flow: Karahan, Pugsley, and Şahin argue that the decline in the growth rate of the US labour force, itself demographic in origin, is a first-order cause of the long decline in the US startup rate — the “demographic origins of the startup deficit” (Karahan, Pugsley, and Şahin 2024). Business dynamism in the United States, measured by firm entry and job reallocation, has declined for decades (Decker, Haltiwanger, Jarmin, and Miranda 2014), and geographic mobility — the willingness to move for opportunity, itself strongly age-related — has fallen substantially (Molloy, Smith, and Wozniak 2011). On the innovation side, Benjamin Jones documents that the age at which inventors do their great work has risen and that innovative output is concentrated in a band of the life-cycle, implying that an older workforce may innovate less per capita (Jones 2010). And a direct study of the question by Nicole Maestas, Kathleen Mullen, and David Powell, using variation in ageing across US states, finds that population ageing did reduce economic growth, with roughly two-thirds of the effect operating through slower productivity growth and one-third through slower labour-force growth (Maestas, Mullen, and Powell 2023). Taken together these are a serious case that ageing dampens dynamism.
But the case is not one-sided, and the contrary evidence is too important to omit. Daron Acemoglu and Pascual Restrepo, examining the cross-country relationship, find no negative association between population ageing and GDP per capita growth over recent decades — if anything a positive one — and argue that this is because ageing induces the adoption of automation technologies that substitute for scarce labour, offsetting the demographic drag (Acemoglu and Restrepo 2017; Acemoglu and Restrepo 2022). This finding directly contradicts the simple ageing-reduces-growth thesis, and by the standard of this essay it cannot be waved away; it must be engaged. The reconciliation, to the extent one is available, is that the two literatures are measuring different things and different margins: Maestas and colleagues exploit within-country variation and find a productivity drag; Acemoglu and Restrepo exploit cross-country variation and find that endogenous technological response can more than compensate, at least where the capacity to automate exists. It is also worth recording a finding that complicates the popular association of youth with innovation: Azoulay, Jones, Kim, and Miranda show that the most successful entrepreneurs are, on average, middle-aged — the founder of a top-performing new firm is far more likely to be around fifty than in their twenties (Azoulay, Jones, Kim, and Miranda 2020). The romantic image of the young disruptor is largely a myth; experience matters for the kind of entrepreneurship that actually generates growth. The defensible conclusion is therefore hedged: there are real mechanisms by which ageing reduces dynamism, and within-country evidence that it has done so, but the aggregate effect is genuinely disputed, the capacity of automation to offset it is real, and the simplest form of the claim — old societies must stagnate — is not supported. What can be said is that ageing removes several of the tailwinds of dynamism, and that whether automation supplies a replacement is contingent, not automatic.
The r > g engine and its counter
There is a further, more theoretical link between ageing and disparity that deserves separate treatment because it is both elegant and contested. Thomas Piketty’s account of long-run inequality rests on the relationship between the rate of return on capital, r, and the growth rate of the economy, g: when r exceeds g, accumulated and inherited wealth grows faster than the economy and than labour incomes, so that the weight of past wealth in the present rises and inequality of wealth tends to widen, with inherited fortunes playing an ever-larger role (Piketty 2014). The demographic connection is direct and often overlooked. The growth rate g has a demographic component — the growth of the labour force — and ageing, by slowing or reversing that growth, lowers g. A lower g, holding r roughly constant, widens the gap r − g, which on Piketty’s account intensifies the dominance of accumulated and inherited wealth. Ageing, on this logic, is not merely correlated with inequality; it is a mechanical driver of the specific, inheritance-based form of inequality that most threatens the ideal of an open, mobile society — and it does so precisely by lowering the growth that would otherwise dilute the weight of old money. This channel also links back to the asset-holder politics of the earlier sections: as inherited housing wealth, concentrated among the old, comes to dominate the distribution, the intergenerational transmission of position strengthens, and the society drifts toward the hereditary character that open-access orders were meant to escape.
The contrary thesis must be given its due, because it is advanced by serious economists and points in the opposite direction. Charles Goodhart and Manoj Pradhan argue, in The Great Demographic Reversal, that the coming decades will invert the pattern of the recent past: as the global working-age population stops growing and begins to shrink — with China’s reversal central to their argument — labour will become scarce relative to capital, which will raise wages, compress the returns to capital, and reduce inequality within countries, even as it raises inflation and interest rates (Goodhart and Pradhan 2020). On this view ageing is, within nations, an equalising force, not a concentrating one, because it restores the bargaining power of labour that the entry of China and the developing world into the global workforce had destroyed. I record this as a genuine and well-argued challenge to the r − g story, not a fringe view, though it is presented in a book rather than in the peer-reviewed journal literature and its central predictions remain, as of writing, largely untested against outcomes. The two theses cannot both be fully right, and which dominates is an open empirical question that the next decade will help settle. The honest position is that ageing has offsetting effects on within-country inequality — a wealth-concentration effect through r − g and the entrenchment of inherited assets, and a potential wage-raising, equalising effect through labour scarcity — and that the net sign is contested rather than known. What is not contested is that ageing acts powerfully on the distribution through both channels; the disagreement is about direction, and I will not resolve by assertion what is genuinely unresolved.
China: old before rich, and closed
China concentrates every thread of this essay into its sharpest form, which is why it belongs at the centre rather than the periphery. It is ageing faster than any large society in history, it is doing so at a far lower level of income than the ageing societies of the West, and it is doing so without the immigration offset on which the West relies. Each of these compounds the others.
The speed and its cause are not in doubt. China’s fertility fell from about 4.45 in 1960 to below replacement by the early 1990s and to roughly 1.01 by 2024 (World Bank, SP.DYN.TFRT.IN) — among the lowest in the world. Part of this fall was the ordinary consequence of development, urbanisation, and rising female education, which lowered fertility across all of East Asia; but part was the coercive one-child policy, in force from around 1979 until its relaxation to two children in 2016 and three in 2021 (Hesketh, Lu, and Xing 2005; Zeng and Hesketh 2016). The relaxations produced almost no rebound, which is itself evidence that the underlying drivers are now structural rather than merely regulatory: the desired family size has fallen, and pro-natal exhortation has not moved it. Cai Yong and others had argued before the policy’s end that China’s fertility had already fallen to a level from which policy could not easily raise it (Cai 2010), and the post-2016 experience has borne this out.
The consequences are arriving in sequence. China’s working-age share peaked around 2010 and its total population around 2021–2022, after which it began to decline — the first sustained decline since the famine of the early 1960s — even as India surpassed it in population (World Bank, SP.POP.TOTL). The old-age dependency ratio has risen from about 7 in 1980 to about 21 in 2025 and is projected to rise steeply for decades. The phrase Chinese demographers themselves use — wei fu xian lao, “growing old before growing rich” — captures the distinctive danger: whereas Japan, Italy, and Germany became old after becoming wealthy, and so met the fiscal burden of ageing from a position of accumulated affluence and mature pension and health systems, China confronts a comparable and faster-arriving burden at a fraction of the income per head and with a far less developed social safety net. The fiscal arithmetic of supporting a rapidly growing elderly population on a shrinking working-age base, at middle-income levels, is severe.
Two further features make China’s case singular. The first is the sex-ratio imbalance produced by the interaction of the one-child policy with son preference and sex-selective abortion: the sex ratio at birth rose well above the natural level through the 1990s and 2000s, producing a cohort of young adults with tens of millions more men than women — the “bare branches” of Valerie Hudson and Andrea den Boer’s analysis, who argued that large surpluses of unmarriageable young men have historically been associated with elevated social instability and can pose security concerns (Hudson and den Boer 2004). This is, in effect, a residual youth-bulge distortion embedded within an ageing society. The second is the absence of an immigration escape valve. Where the West can and does cushion ageing by importing working-age people, China neither receives significant immigration nor conceives of itself as a country that might; its demographic trajectory therefore has no external supplement. A caution is owed on data quality, and it is not a minor one: Chinese demographic statistics have been the subject of sustained expert dispute, with some analysts — Yi Fuxian prominent among them — arguing that official figures have overstated the population and understated the severity of the decline. I flag this as an acknowledged uncertainty rather than resolving it; the broad direction and the rough magnitudes are not seriously contested, but the precise figures carry more uncertainty than those of the OECD democracies.
The political question China poses is genuinely open, and I will pose it as a question rather than answer it by assertion. On one reading, an ageing China is a more stable China: a shrinking share of restive young adults, by the youth-bulge logic, lowers the demographic fuel for the kind of upheaval that has toppled regimes elsewhere, and an older population with more to lose is, by the stakes logic, less inclined to disruption. On the competing reading, an ageing China is a more fragile China: a regime whose legitimacy has rested heavily on delivering rapid growth faces demographic headwinds to that growth precisely as the fiscal demands of an elderly population mount, and the combination of slowing growth, rising pension and health costs, and a shrinking tax base could erode the performance legitimacy on which the system depends. These readings are not mutually exclusive — a regime can be simultaneously insulated against street revolution and strained by fiscal and legitimacy pressures — and which dominates is not something the evidence currently settles. What can be said is that China is running the largest and fastest demographic experiment in modern history, with the fewest of the cushions available to the ageing West, and that the outcome is consequential far beyond its borders.
Europe’s varieties
If China is the extreme case, Europe is the laboratory of variation, and the differences among European states show that demographic outcomes are shaped decisively by two policy-adjacent variables: fertility and the immigration response. Ageing is common to all; its severity and its politics are not.
Germany illustrates the immigration route and its backlash. With fertility around 1.36 and one of the older populations in Europe (about 24% over 65), Germany faced an acute labour and fiscal challenge and responded, most dramatically in 2015, with large-scale admission of refugees and migrants, cushioning its dependency ratio at the cost of a sharp nativist reaction that reshaped its party system. Germany is the cultural-backlash thesis in miniature: the demographic necessity of immigration and the political cost of it, arriving together.
Italy illustrates the trap of low fertility without an adequate offset. Its fertility, around 1.18, is among the lowest in Europe; its share over 65, about 25%, among the highest; and rather than attracting compensating working-age immigration, it has experienced significant emigration of its own young and educated, which accelerates ageing by removing precisely the cohort that would have had children and paid taxes. The combination — very low fertility, net loss of the young, a heavy public-debt burden, and sluggish growth — is close to the pure form of the demographic-stagnation problem, and it shows how ageing, emigration, and fiscal fragility can reinforce one another.
France and Sweden illustrate the mitigating power of higher fertility and, in France’s case, a longer history of immigration. French fertility, around 1.61, and Swedish, around 1.43, are at the upper end of the European range — sustained in part by family policy and in part by immigrant populations — and their dependency ratios, while rising, have risen from a somewhat more favourable base. Neither has escaped the cultural politics of immigration; both show that a less severe demographic problem is compatible with intense political conflict over the means used to soften it. And at the continent’s eastern edge, states such as Bulgaria, Latvia, and Romania display the most brutal version of the Italian pattern: very low fertility combined with heavy emigration of the young to richer EU states, producing some of the fastest population declines in the world and a politics increasingly organised around demographic anxiety and national preservation. The European lesson is that within a shared condition of ageing, the distribution of outcomes is wide, and it is governed substantially by fertility and by the scale and kind of the immigration response — the same two variables that separate China’s trajectory from the West’s.
Demography under the plutocratic ratchet
It is now possible to draw the threads together and to connect this essay to the two that preceded it. The earlier argument was that the United States and Britain have moved into the antechamber of plutocracy: historic concentration of wealth, the legal narrowing of collective resistance, and the hollowing of countervailing power, held short of an outright tip by institutions that still function. The demographic argument supplies the current beneath that drift and explains, in part, why the resistance is weakening at just the moment the concentration is intensifying.
Assemble the mechanisms. An ageing population is, by the compositional and biographical-availability logic, less capable of disruptive collective action, so the demographic fuel of the mass mobilisation that historically forced elites to concede is thinning. An ageing electorate is, by the stakes-and-turnout mechanism and independent of any change in opinion, more oriented toward the defence of accumulated assets and the existing order, so the pivotal voter’s interest increasingly runs with the grain of concentration rather than against it. Ageing lowers g and, through the r − g channel, tends to entrench inherited and accumulated wealth — the housing wealth of the old above all — deepening the specific hereditary form of inequality that open societies exist to prevent, even as a serious contrary thesis holds that labour scarcity may push the other way. And immigration, the one remedy for the economics of ageing, generates a cultural backlash concentrated in the ageing native population — a backlash that populist entrepreneurs can and do direct toward cultural and national targets while leaving the distribution of wealth intact, so that the discontent of the demographically anxious is channelled away from concentration rather than against it. Each of these lowers, in its own way, the resistance that a drift toward plutocracy would otherwise meet. This is what it means to call demography the current beneath the politics: it does not command the outcome, but it sets the odds, and it is setting them against the countervailing power on which the earlier essays’ hopes rested.
The link is a matter of tendency and probability, not of mechanism-free destiny, and I hold it at that level deliberately. Demography does not repeal politics; mobilisation can still occur, institutions can still hold, and the contrary economic theses may prove right. But the configuration is unfavourable, and it is unfavourable in a mutually reinforcing way: the same ageing that quiets the population also entrenches the assets, and the same immigration that could rebuild the tax base also ignites the backlash that protects the concentration. The plutocratic ratchet of the earlier essays turns more easily in an old society, because an old society supplies less friction at every point where friction was supposed to arrest it.
Counterarguments and limits
The standard of this sequence requires stating the strongest objections at full strength, and several bear directly on the argument just made.
The cohort objection. The claim that ageing tilts politics toward the status quo rests, in its psychological form, on shaky ground, and I have conceded as much. If the political colour of the old is set by the formative experiences of their youth rather than by ageing itself, then the politics of tomorrow’s elderly will be the politics of today’s young, and no general law of gerontocratic conservatism holds. My argument survives this objection only because it does not depend on the psychological claim: the stakes-and-turnout mechanism operates through the distribution of assets and the propensity to vote, not through attitude change, and it holds even if every individual’s opinions are fixed for life. But the reader should be clear that the psychological version of “ageing makes societies conservative” is not something I am asserting, because it is not something the evidence supports.
The dynamism objection. Acemoglu and Restrepo’s finding that ageing has not reduced growth cross-nationally, because automation compensates, is a genuine challenge, and I have not resolved it. If they are right at the aggregate level, then the “less dynamism” strand of the popular demographic pessimism is overstated, and ageing societies may sustain growth through capital deepening even as their workforces shrink. My position is only that the mechanisms reducing dynamism are real and that within-country evidence supports a productivity drag, while conceding that the offsetting technological response is also real and that the net aggregate effect is disputed.
The inequality-direction objection. Goodhart and Pradhan’s thesis that ageing will reduce within-country inequality by making labour scarce is the most direct challenge to the disparity strand of the argument, and it is advanced by serious people. I have therefore stated the r − g channel and the labour-scarcity channel as offsetting, with the net sign contested rather than known. Anyone claiming certainty that ageing raises inequality is, on the current evidence, overclaiming — as is anyone certain it lowers it.
The peace objection. One might grant the whole argument and ask why a less rebellious, more stable society is a bad thing — is quiet not preferable to upheaval? The answer, as in the earlier essays, is that peace is not the same as freedom or fairness. A society can be stable because its arrangements are just and broadly consented to, or stable because the demographic and economic capacity to contest them has withered. The concern here is the second kind of stability: not the peace of a settled and open order, but the quiet of a population that has lost the means to push back. That such a society does not revolt is not evidence that it has nothing to revolt against.
The data-quality objection. The demographic figures for the OECD democracies are among the most reliable social statistics that exist, but the Chinese figures, on which part of the argument leans, carry real and acknowledged uncertainty, and the long-run projections for every country are model outputs, not observations. I have tried to lean on the recorded past rather than the projected future, and to flag the Chinese case as more uncertain than the rest; the broad directions are secure, but precision beyond them is not claimed.
Conclusion
Demography is not destiny, and I have tried throughout to hold the argument to that standard rather than to overrun it. What demography is, is a current — a slow, deep force that changes the probabilities of political and economic outcomes without dictating them, and that is more powerful for being largely invisible, unfolding over decades rather than announcing itself in events. The current now runs, across the developed world and most sharply in China, toward societies that are older, and older societies are — for reasons of composition and stakes rather than any mystical property of age — less prone to the disruptive collective action that has historically forced concentrated power to yield. They are more oriented, through the arithmetic of turnout and the distribution of assets, toward the defence of the existing order, whether or not a single mind has changed. They may be less dynamic, though that is genuinely contested and automation may compensate. And they are subject to offsetting pressures on inequality whose net direction the coming decade will decide, but which act, either way, to make the distribution of wealth a more central and more fraught political question than it has been for generations.
Immigration sits at the hinge of all of it: the one available offset to the economics of ageing, and the one most certain to provoke a backlash among the ageing populations that most need it — a backlash that leaves the concentration of wealth not merely intact but, frequently, better protected, because it redirects discontent toward the foreigner and away from the fortune. And China stands as the case that concentrates every element into its starkest form: ageing faster than any large society ever has, at a lower income than any of its ageing peers, with none of the cushions the West relies on, and with a political system whose response to the resulting strain the world has reason to watch closely.
The earlier essays asked whether the United States and Britain were becoming plutocracies and answered: not yet, but they stand in the antechamber, with the door narrowing. This essay identifies part of what is narrowing it. An ageing, asset-holding, less mobilised electorate, in a society whose dynamism is uncertain and whose inequality is entrenching through the weight of inherited wealth, is a society in which the friction that once arrested the drift toward rule by wealth is quietly wearing away. The weight of years does not push the door shut. It removes the hands that were holding it open.
Works Cited
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., & 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.
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.
Borjas, G. J. (2017). The Wage Impact of the Marielitos: A Reappraisal. ILR Review, 70(5), 1077–1110.
Cai, Y. (2010). China’s Below-Replacement Fertility: Government Policy or Socioeconomic Development? Population and Development Review, 36(3), 419–440.
Card, D. (1990). The Impact of the Mariel Boatlift on the Miami Labor Market. ILR Review, 43(2), 245–257.
Card, D. (2009). Immigration and Inequality. American Economic Review, 99(2), 1–21.
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.
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.
Goldstone, J. A. (2002). Population and Security: How Demographic Change Can Lead to Violent Conflict. Journal of International Affairs, 56(1), 3–21.
Goodhart, C., & Pradhan, M. (2020). The Great Demographic Reversal: Ageing Societies, Waning Inequality, and an Inflation Revival. Palgrave Macmillan.
Hainmueller, J., & Hiscox, M. J. (2010). Attitudes toward Highly Skilled and Low-Skilled Immigration: Evidence from a Survey Experiment. American Political Science Review, 104(1), 61–84.
Hainmueller, J., & Hopkins, D. J. (2014). Public Attitudes Toward Immigration. Annual Review of Political Science, 17, 225–249.
Hesketh, T., Lu, L., & Xing, Z. W. (2005). The Effect of China’s One-Child Family Policy after 25 Years. New England Journal of Medicine, 353(11), 1171–1176.
Hudson, V. M., & den Boer, A. M. (2004). Bare Branches: The Security Implications of Asia’s Surplus Male Population. MIT Press.
Inglehart, R., & Norris, P. (2017). Trump and the Populist Authoritarian Parties: The Silent Revolution in Reverse. Perspectives on Politics, 15(2), 443–454.
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.
Krosnick, J. A., & Alwin, D. F. (1989). Aging and Susceptibility to Attitude Change. Journal of Personality and Social Psychology, 57(3), 416–425.
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. (1952). The Problem of Generations. In Essays on the Sociology of Knowledge (pp. 276–322). Routledge & Kegan Paul. (Original work published 1928.)
McAdam, D. (1986). Recruitment to High-Risk Activism: The Case of Freedom Summer. American Journal of Sociology, 92(1), 64–90.
Molloy, R., Smith, C. L., & Wozniak, A. (2011). Internal Migration in the United States. Journal of Economic Perspectives, 25(3), 173–196.
Norris, P., & Inglehart, R. (2019). Cultural Backlash: Trump, Brexit, and Authoritarian Populism. Cambridge University Press.
Peri, G., & Yasenov, V. (2019). The Labor Market Effects of a Refugee Wave: Synthetic Control Method Meets the Mariel Boatlift. Journal of Human Resources, 54(2), 267–309.
Piketty, T. (2014). Capital in the Twenty-First Century (A. Goldhammer, Trans.). Harvard University Press.
Ryder, N. B. (1965). The Cohort as a Concept in the Study of Social Change. American Sociological Review, 30(6), 843–861.
Sears, D. O., & Funk, C. L. (1999). Evidence of the Long-Term Persistence of Adults’ Political Predispositions. Journal of Politics, 61(1), 1–28.
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.
United Nations Population Division. (2000). Replacement Migration: Is It a Solution to Declining and Ageing Populations? United Nations.
Urdal, H. (2006). A Clash of Generations? Youth Bulges and Political Violence. International Studies Quarterly, 50(3), 607–629.
Zeng, Y., & Hesketh, T. (2016). The Effects of China’s Universal Two-Child Policy. The Lancet, 388(10054), 1930–1938.
Data source: World Bank, World Development Indicators (drawing on the UN World Population Prospects and national statistical offices): total fertility rate (SP.DYN.TFRT.IN), population aged 65+ as % of total (SP.POP.65UP.TO.ZS), old-age dependency ratio (SP.POP.DPND.OL), working-age population as % of total (SP.POP.1564.TO.ZS), and total population (SP.POP.TOTL), retrieved 2025. US and UK voting-by-age patterns: US Census Bureau Current Population Survey voting supplements; British Election Study.