The Global Review - Farid Novin

The Socioeconomic Transformation of Mature

Democracies: Europe and the United States

Demographic Change, Economic Inequality, Automation and Artificial Intelligence, Energy and Price Shocks, Information Systems and the Strain on Democratic Legitimacy

A Comparative Socioeconomic, Geoeconomic and Bayesian Game-Theoretic Assessment as of November 10, 2026

Farid Novin

 

Abstract

This assessment examines the interacting socioeconomic pressures that are testing the institutions of mature democracies in Europe and the United States, using information published on or before November 10, 2026. It argues that the present strain on democratic legitimacy is best understood as a multidimensional institutional adjustment crisis rather than the product of any single cause. Six transformations are operating at once: demographic change, a widening divide between asset owners and wage earners, the diffusion of automation and artificial intelligence, the fragmentation of the information environment, an energy and price shock arising from the war involving Iran and the disruption of the Strait of Hormuz, and a declining capacity of parties, legislatures and administrations to respond at the speed that events demand.

The evidence assembled here supports four principal findings. First, the economic pressures are real but unevenly distributed: the share of United States household net worth held by the wealthiest one per cent reached a record 32.5 per cent in the second quarter of 2026, while 92.7 million people in the European Union, or 20.9 per cent of the population, were at risk of poverty or social exclusion in 2025. Second, the labour-market effects of artificial intelligence are, so far, concentrated rather than economy-wide: payroll data show a widening employment gap for workers aged 22 to 25 in highly exposed occupations, yet there is no evidence of general displacement, and competing explanations such as remote work remain credible. Third, the information environment has become a source of democratic anxiety on both sides of the Atlantic, with a median of 68 per cent across 37 surveyed countries saying that the internet and social media have made people more politically divided. Fourth, the autumn of 2026 has produced two natural experiments in institutional resilience, the French budget crisis and the United States midterm elections, whose outcomes point in different directions and caution against deterministic readings of either region’s trajectory.

The paper situates the political-violence literature, including the work of Robert Pape, within a wider family of explanations, and develops a conceptual Bayesian game-theoretic framework in which the probability of democratic instability depends on the interaction of structural pressure, institutional mediation and elite strategy. It closes with policy implications for the distribution of technological gains, housing, education, migration and integration, democratic information infrastructure and public administration.

I. Introduction: Beyond Single-Factor Explanations

Few questions in comparative politics are more pressing than why established democracies with high incomes, long constitutional traditions and strong civil societies appear increasingly fragile. The scholarly literature offers several families of explanation, and each captures something important.

One family emphasises relative deprivation and status threat. Ted Robert Gurr’s classic argument in Why Men Rebel (1970) held that political violence tends to arise where the gap between what people expect and what they believe they can obtain widens. More recent work on cultural backlash, notably by Pippa Norris and Ronald Inglehart (2019), links support for populist movements to perceived threats to the status of formerly dominant social groups. Robert Pape’s analysis of violent populism in Our Own Worst Enemies: America in the Age of Violent Populism (2026) belongs to this tradition, stressing demographic change, the concentration of wealth, polarising rhetoric and the normalisation of political violence.

A second family stresses economic geography and the distributional shocks of trade and technology. David Autor, David Dorn and Gordon Hanson’s work on the China trade shock (2013) showed that local labour markets absorb adjustment unevenly and with lasting scars. Dani Rodrik (2018) has argued that the populist backlash can be read as a political response to the distributional consequences of globalisation. Anne Case and Angus Deaton (2020) documented the human cost of the erosion of working-class prospects in the United States. The OECD’s Employment Outlook 2026, which is devoted to geographic disparities, extends this line of analysis to the age of artificial intelligence.

A third family concentrates on elites and institutions. Steven Levitsky and Daniel Ziblatt (2018) argue that democracies die less through dramatic coups than through the gradual abandonment of the norms of mutual toleration and institutional forbearance by political leaders. Barbara Walter (2022) identifies the partial democracy, in which institutions are weakening but not yet closed, as the regime type most vulnerable to civil conflict. Daron Acemoglu and James Robinson (2019) emphasise the balance between state capacity and an organised society.

A fourth family treats technology as a question of power and direction rather than inevitability. Acemoglu and Simon Johnson (2023) argue that the social consequences of technological change depend on choices about whether innovation augments workers or replaces them, and on who captures the gains. Erik Brynjolfsson and colleagues at the Stanford Digital Economy Lab have produced some of the most granular evidence yet on how generative artificial intelligence is affecting the early careers of young workers.

These families are complementary rather than competing. The central proposition of this assessment is that the present crisis of mature democracies is not simply a demographic transition accompanied by rising inequality, nor merely a failure of elite norms, nor solely a technological shock. It is a multidimensional institutional adjustment crisis arising from the simultaneous transformation of population structures, the distribution of wealth and income, productive technologies, information systems, the global energy and price environment, and the capacity of political authority to respond.

The distinction matters for policy. If governments interpret the crisis exclusively through immigration, cultural identity or redistribution, they risk neglecting the technological, informational and institutional changes that are reshaping employment, social mobility and democratic participation. Conversely, an exclusively technological reading would overlook the demographic and distributional politics that determine how technological gains are shared.

The analysis that follows proceeds comparatively. It distinguishes structural economic pressures, their political consequences and the mechanisms through which pressure may or may not translate into democratic instability. It also treats cross-national evidence with caution: surveys measure attitudes rather than behaviour, aggregate indicators can mask regional concentration, and several of the relationships discussed below are hypotheses to be tested rather than established causal effects.

II. Six Interacting Transformations

The most useful way to organise the evidence is to distinguish six transformations that are occurring simultaneously and that reinforce one another.

The first is demographic transformation: population ageing, low fertility, migration, changing ethnic and cultural composition, and the geographic redistribution of population within countries. The second is distributional transformation: the concentration of wealth, housing unaffordability, declining relative mobility and the divergence between owners of appreciating assets and people who depend on wages. The third is technological transformation: automation, robotics, digital platforms and, above all, generative artificial intelligence, which extends automation from routine physical tasks to parts of cognitive and professional work. The fourth is informational transformation: the internet, social media, algorithmic recommendation, synthetic media and the decline of shared sources of news. The fifth is macroeconomic and geopolitical transformation: in 2026 this chiefly means the energy and price shock associated with the war involving Iran and the closure of the Strait of Hormuz, together with fiscal strain, higher interest rates and trade tension. The sixth is institutional transformation: the declining capacity of parties, legislatures, central banks, public administrations and multilevel governance structures to respond at the speed and scale that these other changes require.

The sixth transformation is the mediating one. The same demographic, technological or price shock can produce very different political outcomes depending on whether institutions are able to absorb and distribute its costs. This is the reason that comparisons between Europe and the United States are illuminating: the two regions face similar pressures but have different institutional endowments.

The United States combines a more dynamic technology sector, comparatively flexible labour markets, very high wealth concentration, thinner social insurance in healthcare and income protection, and a highly polarised two-party system. Europe combines more extensive social protection, older population structures, different migration histories, greater institutional fragmentation, and a complicated relationship between national sovereignty and supranational governance. Both face a common difficulty: political and administrative institutions designed for slower economic and social conditions are being asked to manage changes that are moving much faster than institutions usually adapt.

III. The Macroeconomic and Geopolitical Backdrop, November 2026

Any assessment of democratic strain in November 2026 must begin with the energy and price environment, because it has altered the political economy of both regions during the year.

1. Global growth, concentration and risk

The International Monetary Fund’s October 2026 World Economic Outlook, titled Resilience Under Strain, Urgent Choices, projects global growth of 3.1 per cent in 2026, down from 3.5 per cent in 2025, and 3.4 per cent in 2027. United States growth is projected at 2.3 per cent in both years, supported by consumption, fiscal policy and technology investment, while the euro area is projected to grow by 1.1 per cent in 2026 and 1.2 per cent in 2027, held back by high energy prices and weak consumer confidence. The Fund reports that growth has become more concentrated: just 32 economies now account for 95 per cent of global growth, compared with 54 in its April assessment, and roughly 47 per cent of economies had their forecasts downgraded.

The Fund’s baseline assumes that the Strait of Hormuz reopens in March 2027, with conditions broadly returning to pre-war levels by October 2027. Before the war the strait carried about a quarter of the world’s seaborne oil trade. The Fund projects global inflation of 4.7 per cent in 2026, easing to 4.1 per cent in 2027, and expects core inflation not to return to 2 per cent until mid-2028 in the euro area and the United States. It identifies geopolitical conflict, fiscal vulnerability, trade tension and a possible reassessment of expectations surrounding artificial intelligence as the principal downside risks, and it notes that stabilising public debt ratios in most large economies would require more fiscal consolidation than current policies envisage.

2. The United States: inflation, monetary tightening and a resilient labour market

In the United States, the Bureau of Labor Statistics reported that consumer prices rose 3.7 per cent over the twelve months to September 2026, up from 3.4 per cent in August, with the energy index up 20.4 per cent over the year and the gasoline index up 9.1 per cent in September alone. Core prices, which exclude food and energy, rose 0.3 per cent in each of August and September, and the shelter index rose 3.2 per cent over the year. For October, the Bureau reported a monthly increase of 0.4 per cent, above forecasts; because the 2025 federal government shutdown left a gap in October 2025 price collection, no annual rate was published for that month.

The Federal Reserve, under Chair Kevin Warsh, raised the federal funds target range by 25 basis points to 3.75 to 4.00 per cent on September 16, the first increase since July 2023, and held the range on October 28 by a vote of 10 to 2, with Governor Lisa Cook and Cleveland Fed President Beth Hammack dissenting in favour of a further quarter-point increase. The Committee’s next meeting is on December 8 and 9. At the October meeting the Chair also announced a change in the institutional architecture of policy: from 2027 one meeting a year will be devoted to strategy rather than rate-setting, moving toward a schedule of six rate-setting meetings and two strategy meetings.

The labour market has proved more resilient than many forecasters expected. Nonfarm payrolls rose by 165,000 in October, well above consensus, after an upward revision of September to 60,000, and the unemployment rate fell to 4.1 per cent, within the narrow 4.1 to 4.3 per cent range that has prevailed since March. Labour force participation was 61.9 per cent and the employment-to-population ratio 59.3 per cent. Average hourly earnings rose by 0.4 per cent in October to $38.00. Yet the broader U-6 measure, which includes discouraged workers and people working part time for economic reasons, rose from 7.6 to 7.9 per cent. The headline strength and the pockets of underemployment illustrate a theme developed later: aggregate indicators can look sound while the distribution of opportunity deteriorates.

3. The euro area: an energy-driven inflation shock and a hiking central bank

In the euro area, Eurostat’s flash estimate put annual inflation at 4.0 per cent in October 2026, the highest in over three years, up from 3.8 per cent in September and 3.2 per cent in August. Energy prices were 20.8 per cent higher than a year earlier, services inflation was 3.3 per cent and core inflation 2.5 per cent. The European Central Bank raised its deposit facility rate by 25 basis points to 2.50 per cent in September, held it at its October meeting, and faces a decision on December 17 in an environment in which several policymakers have signalled support for a further increase. Flash purchasing managers’ indices for October indicated slowing growth, which places the Bank in the difficult position of tightening policy into a weakening economy.

The political significance of these figures is considerable. An energy-driven price shock is felt immediately and visibly in household budgets, fuels, heating and food, and it arrives on top of the cumulative price increases of the preceding years. It also compresses the fiscal space that governments need to cushion households, at precisely the moment when ageing, defence, climate and technology investment are competing for the same resources.

IV. Demographic Transformation

Demographic change is frequently invoked as a source of political tension, but it comprises three separate processes with different economic and political implications: population ageing, migration, and changes in ethnic and cultural composition. Treating them as one phenomenon obscures more than it reveals.

1. Europe: ageing, negative natural change and dependence on migration

According to Eurostat, the European Union had an estimated 452.0 million inhabitants on January 1, 2026, an increase of 706,000 over the previous year and the fifth consecutive year of growth following the decline of 2021. Natural change has, however, been negative every year since 2012: in 2025 deaths exceeded births by roughly 1.35 million, and the overall increase was driven by net migration of about 2 million. Population growth was highly uneven. It increased in 16 member states and fell in 11, with Spain alone accounting for roughly two-thirds of the Union’s net increase, while Latvia, Estonia, Hungary and Bulgaria recorded some of the sharpest declines.

This creates a structural dilemma. An ageing population increases demand for pensions, healthcare and long-term care, while the relative size of the working-age population that finances them is shrinking in much of the Union. Immigration can moderate labour shortages and demographic contraction, but its economic and political success depends on integration infrastructure: housing, schooling, language acquisition, recognition of professional qualifications, access to employment and administrative capacity. Where these systems are insufficient, immigration becomes a focal point for political competition, and the resulting conflict is not confined to citizens and newcomers. It can also pit younger against older cohorts, urban against rural populations, taxpayers against recipients of public benefits, and established workers against new entrants.

2. The United States: slower growth, ageing and a changing composition

The Congressional Budget Office’s January 2026 demographic projections show the United States population growing from 349 million in 2026 to 364 million in 2056, at an average rate that slows from 0.3 per cent a year over the next decade to 0.1 per cent thereafter. Fertility is projected at about 1.58 births per woman in 2026, falling to 1.53 by 2036, well below the replacement level of 2.1. From 2030 annual deaths are projected to exceed births, so that net immigration accounts for all population growth; without immigration, the population would begin to shrink in 2030. The Office also lowered its near-term immigration projections because of administrative actions, the 2025 reconciliation act and revisions to past data, and it stresses that its fertility, mortality and immigration projections are highly uncertain.

The composition of the population has also changed substantially. According to Census Bureau data, the non-Hispanic white share of the population fell from about 76 per cent in 1990 to about 58 per cent in the 2020 census, and it has continued to decline in subsequent estimates. Pape and others interpret this transition as a source of political anxiety among some constituencies, consistent with the status-threat literature. The economic significance, however, lies in how compositional change interacts with labour markets, education, housing and wealth accumulation. A society can diversify while preserving, or widening, inequalities between groups and generations, and political dissatisfaction should not be attributed to demographic change alone. Some voters interpret changing composition as a threat to cultural or political standing, while others experience economic insecurity arising from stagnant real earnings, housing costs or declining employment prospects. These perceptions become politically reinforcing when parties transform legitimate disagreements about migration, public spending and opportunity into existential conflicts over national identity.

3. The demographic paradox

Both societies face a paradox. Ageing democracies require productive workers, new investment and a sustainable tax base, yet the political debate frequently treats migration, technological change and public expenditure as separate ideological questions. They are economically interconnected. A country that restricts migration while also failing to raise productivity will find its fiscal position deteriorating as the dependency ratio rises, while a country that admits migrants without investing in housing and integration will encounter political resistance that eventually undermines the policy. Neither region can resolve its demographic difficulties through restriction alone, and neither can rely on migration alone to solve its productivity and fiscal challenges.

An important qualification follows. Demographic diversity does not automatically generate political violence or democratic erosion. The political consequences depend on institutional responses, economic opportunity, residential integration and the conduct of political elites. The evidence for this qualification is that many demographically diverse and ageing societies remain stable, and that political entrepreneurs, rather than demography itself, choose whether to frame change as threat.

V. Distribution, Wealth and Housing

Contemporary inequality is increasingly a matter of ownership of appreciating assets rather than annual earnings alone. The distinction is fundamental. An employee depends primarily on wages, while a household that owns property, financial securities, businesses or productive capital benefits from asset appreciation, dividends, rents and capital gains. When returns to asset ownership persistently exceed wage growth, as Thomas Piketty argued in Capital in the Twenty-First Century (2014), the distance between owners and non-owners widens even when employment is strong.

1. The United States: record concentration in an equity-driven boom

The Federal Reserve’s Distributional Financial Accounts, released on September 18, 2026, show that the wealthiest one per cent of households held 32.5 per cent of total household net worth in the second quarter of 2026, a record in the series and up from 31.7 per cent in the first quarter. The top 0.1 per cent held 15.0 per cent, while the bottom half of households held about 2.3 per cent. The Federal Reserve’s Financial Accounts, published on September 11, 2026, reported that household and nonprofit net worth rose by a record $12.8 trillion in the second quarter to $195.9 trillion, driven primarily by capital gains on corporate equities.

These figures describe shares, not levels, and they are published roughly a quarter in arrears; they do not by themselves show that the bottom half has become poorer. Nonetheless they illustrate why distribution is politically consequential. Financial assets provide access to investment returns and a capacity to absorb shocks, and equity-led booms disproportionately benefit those who already own equities. Households without meaningful reserves are more exposed to unemployment, medical expenses, interest-rate increases and housing-market disruption, and the current environment of energy-driven inflation and rising policy rates intensifies that exposure.

Housing compounds the problem. For younger Americans, access to homeownership increasingly depends on family assistance, inherited wealth or exceptionally well-paid employment. The shelter component of the consumer price index rose 3.2 per cent over the year to September 2026. The resulting frustration is not confined to low-income households; it increasingly affects middle-income professionals who are employed but cannot acquire the assets traditionally associated with middle-class security.

A distinction must be preserved between perceived and measured decline. Political resentment can grow even when real incomes are improving, if expectations, relative opportunities and access to assets are deteriorating. The October jobs data, with unemployment at 4.1 per cent, illustrate that employment can be strong while affordability and asset access remain the focus of grievance.

2. Europe: social protection, regional divides and a housing squeeze

European welfare states provide broader social insurance than the United States, and their healthcare, unemployment support and public services moderate the consequences of income inequality. These protections do not, however, eliminate inequalities in housing, financial wealth, educational opportunity or the intergenerational transmission of assets.

Eurostat reported that 92.7 million people in the European Union, equivalent to 20.9 per cent of the population, were at risk of poverty or social exclusion in 2025, a decrease of about 600,000 from 2024. The risk was highest in Bulgaria at 29.0 per cent, Greece at 27.5 per cent and Romania at 27.4 per cent, and lowest in Czechia at 11.5 per cent, Poland at 15.0 per cent and Slovenia at 15.5 per cent. Children remain particularly exposed: 24.3 per cent, or 19.3 million, were at risk. Eurostat’s regional analysis, published in October 2026, found that in 24 of the 243 regions for which data were available at least 31.5 per cent of the population was at risk, with the highest share, 54.8 per cent, in the French outermost region of Guyane. The Union’s 2030 target is to reduce the number of people at risk by at least 15 million, including at least 5 million children.

Housing is particularly important. According to Eurostat data reported in October 2026, 7.7 per cent of the EU population lived in households whose housing costs exceeded 40 per cent of disposable income in 2025, with Greece at 26.4 per cent, Denmark at 14.7 per cent, Germany at 11.2 per cent and Sweden at 10.8 per cent. Germany had the largest absolute number affected, at 9.4 million people. Among people at risk of poverty, 29.1 per cent in the Union faced housing cost overburden. Earlier Eurostat data for 2024 showed that young people aged 15 to 29 were more affected than the population as a whole, at 9.7 per cent against 8.2 per cent, with the gap widest in Denmark and the Netherlands, and that young Europeans left the parental home at an average age of 26.3 in 2025. Eurostat has also reported that house prices in the Union rose by 69 per cent on average between 2015 and the second quarter of 2026.

Housing is not simply another consumption expense. It determines where people can live, whether they can form families, how far they must travel to work and whether they can accumulate wealth. Where younger generations conclude that homeownership and economic independence are becoming inaccessible, the intergenerational social contract becomes fragile, and the political demand for change grows regardless of the aggregate poverty rate.

3. The common transatlantic problem: the asset-ownership divide

The United States and Europe differ in their welfare arrangements, but both confront the same distributional problem: economic growth does not automatically translate into broadly shared economic security. A society can experience rising corporate profits, financial-market appreciation and technology investment while substantial sections of the population experience deteriorating affordability. This is the asset-ownership divide, and it is a natural extension of the relative-deprivation literature: people can perceive economic progress as something that accrues to others while leaving their own circumstances unchanged.

The OECD’s Employment Outlook 2026 adds a geographic dimension. It documents large and persistent gaps in employment, unemployment and disposable income across regions within countries, and concludes that regional access to job opportunities plays a key role in both income levels and mobility. It also finds that mobility across regions alone is insufficient to close these gaps and can sometimes widen them, and that adjustment often occurs through transitions into joblessness and the entry of new workers rather than through displaced workers moving between sectors, leaving lasting scars. The political implication is that national averages understate the concentrated grievance in particular places.

VI. Automation and Artificial Intelligence: Evidence, Uncertainty and Distribution

Earlier waves of automation primarily affected repetitive physical work, assembly-line production and routine clerical processes. Generative artificial intelligence is different because it can perform portions of tasks that were previously considered dependent on human judgement and professional training: drafting and editing text, summarising legal and financial documents, writing and debugging software, handling customer enquiries, producing marketing content and analysing data. The implications therefore extend well beyond manufacturing into the educated middle class.

The analytical distinction that matters is between automating an occupation and automating tasks within it. An accountant who uses artificial intelligence to process information more quickly is not replaced; a law firm may need fewer junior researchers while retaining senior lawyers; a manufacturer may expand output through robotics without a proportional increase in staff. The central question is therefore not whether artificial intelligence will eliminate employment, but whether its productivity gains will be accompanied by sufficient job creation, retraining, wage growth and broad access to the ownership of the resulting productive capital.

1. What the labour-market evidence shows so far

The most granular evidence currently available comes from the Stanford Digital Economy Lab. Its August 2026 revision of Canaries in the Coal Mine? Six Facts About the Recent Employment Effects of Artificial Intelligence, by Erik Brynjolfsson, Bharat Chandar and Ruyu Chen, uses payroll records through June 2026. It reports no widespread, economy-wide displacement associated with artificial intelligence. It does find, however, that employment among workers aged 22 to 25 in the most highly exposed occupations is now about 19 per cent below where it would be had it kept pace with similarly aged workers in less exposed occupations, up from 15 per cent in the July 2025 data. In levels, employment of workers in this age group in the two most exposed quintiles of occupations fell by about 11 per cent between November 2022 and June 2026, while employment of the same age group in the three least exposed quintiles grew by about 10 per cent. Experienced workers show no comparable gap.

This finding is important but must be read carefully. A 19 per cent gap does not mean that 19 per cent of entry-level jobs have disappeared; it measures a shortfall relative to a benchmark. Other research offers competing or complementary explanations. Analysts at the Federal Reserve Bank of New York estimated in June 2026 that remote work can account for about 64 per cent of the rise in unemployment among young college graduates between 2017–2019 and 2022–2024, because firms find it harder to train and mentor newcomers at a distance, and they note that the rise began before generative artificial intelligence spread. The New York Fed’s quarterly tracker put unemployment among recent graduates aged 22 to 27 at about 5.6 per cent in the second quarter of 2026, above the national rate, with about 42 per cent underemployed in jobs that typically do not require a degree. Robert Fairlie and Jane Wu of the University of California, Los Angeles, using household survey data for June to August 2026, found that unemployment among recent graduates did not spike over the summer.

The reasonable reading of this evidence is that artificial intelligence is one contributor, alongside remote work, post-pandemic hiring adjustments, interest rates and cyclical conditions, to a narrowing of the first rung of professional careers, and that the effect is concentrated in occupations where the technology automates rather than assists. The risk to the traditional pathway from education to a first professional post, experience, rising earnings and eventual independence is real, even though economy-wide unemployment remains low. The United States unemployment rate of 4.1 per cent in October 2026 and the stable aggregate employment of heavily exposed occupations are consistent with a transition that is gradual, concentrated and not yet visible in headline statistics.

2. Four distributional channels

There are at least four channels through which artificial intelligence could affect distribution. The first is the displacement or thinning of routine and semi-routine employment in administrative, clerical and customer-facing occupations. The second is pressure on professional and middle-income work, which may spread economic insecurity from traditional industrial working classes to graduates and administrative professionals. The third is the concentration of returns: ownership of computing infrastructure, advanced models, cloud platforms, proprietary data and intellectual property can generate substantial advantages, and if those assets remain concentrated among a small number of firms and investors, technological progress may raise aggregate productivity while widening wealth disparities. This is particularly significant in the United States, where the technology sector is closely tied to capital markets, and the record equity-driven rise in top-end wealth documented above illustrates the link. The fourth is geography. The OECD finds that trade and technology shocks, including artificial intelligence, are reshaping local labour markets, with some regions losing manufacturing jobs while others create service and non-routine work, and that displaced workers rarely transition into the growing sectors even when these are nearby. A technology firm may create thousands of high-value jobs in a metropolitan centre while automation erodes employment in manufacturing towns; the national economy may benefit, but the political consequences of the geographic distribution may be destabilising.

Daron Acemoglu (2025) has offered a more sceptical macroeconomic assessment of near-term aggregate gains, arguing that productivity effects may be more modest than the most enthusiastic projections, while Acemoglu and Johnson (2023) emphasise that outcomes depend on whether technology is directed toward augmenting workers or replacing them. The Fund’s October 2026 assessment points in a complementary direction: it identifies strong artificial intelligence investment as a source of upside to growth, but warns that a reassessment of expectations could trigger financial-market adjustment. A technology investment boom should therefore not be treated as proof of an inclusive productivity transformation. The distinction between financial-market expectations and realised productivity will matter increasingly for monetary policy, fiscal planning and financial stability.

3. The European adoption divide

Europe faces an additional challenge in the unevenness of adoption. Eurostat reports that 20.0 per cent of EU enterprises with ten or more employees used artificial intelligence technologies in 2025, up from 13.5 per cent in 2024 and 8.1 per cent in 2023. Adoption is strongly related to firm size: 55 per cent of large enterprises used these technologies, compared with 19 per cent of small and medium-sized enterprises. It is also strongly related to geography, ranging from 42.0 per cent in Denmark, 37.8 per cent in Finland and 35.0 per cent in Sweden to 5.2 per cent in Romania, 8.4 per cent in Poland and 8.5 per cent in Bulgaria. Large firms can finance specialised computing infrastructure, recruit engineers and reorganise operations, whereas smaller firms frequently face financial, technical and skills constraints; the OECD’s December 2025 report on small-business adoption, prepared for the Group of Seven, found that about half of surveyed firms cited skills shortages as their primary barrier.

There is also a human-capital problem. Eurostat’s 2026 digitalisation publication reports that 40 per cent of EU citizens lacked basic or above-basic digital skills in 2025, even though 93 per cent used the internet at least weekly. Europe therefore risks rising demand for advanced digital competencies alongside a large population that cannot fully participate in the new economy, and a widening productivity divide between large firms and the small enterprises that account for much of its employment.

4. The American paradox

The United States possesses considerable advantages in artificial intelligence research, advanced computing, venture capital and software development, and the International Monetary Fund attributes part of its resilient growth to technology investment. Technological leadership does not, however, guarantee broadly distributed prosperity. The American paradox is that the economy may become more productive and technologically sophisticated while parts of the workforce, notably young professionals entering exposed occupations, experience weaker bargaining power and less secure career ladders. The issue is not only unemployment; it is the potential erosion of the pathways through which individuals have historically entered the middle class.

5. Productivity without inclusion

If artificial intelligence increases output while reducing the labour required to produce it, the gains can be distributed through lower prices, higher wages, shorter working hours, higher profits, lower taxation or expanded public services. There is no automatic mechanism ensuring that these channels operate in a balanced way. If the principal beneficiaries are the owners of technology and financial capital, the result will be a more productive but less inclusive economy. The proposition advanced here is that the next major economic divide may be less a divide between capital and labour in the traditional sense than a divide between those who own, control and effectively use advanced technologies and those whose economic position is weakened by them. This is not inevitable. It is a question of institutional design, competition policy, education, investment and the distribution of productivity gains.

VII. The Internet and Algorithmic Information Systems

Explanations of political instability that emphasise elite rhetoric must also ask how that rhetoric now reaches the public, acquires credibility and mobilises action. The infrastructure of public communication has changed fundamentally with the internet, social media and algorithmic distribution, and this is not simply a communications revolution but a transformation in the organisation of public opinion.

1. The decline of a shared information environment

For much of the twentieth century newspapers, radio and television provided identifiable sources of public information. They were never neutral and their audiences were never homogeneous, but a limited number of major institutions helped to establish a common framework for national debate. The internet has expanded democratic opportunity: citizens can obtain information from thousands of sources, communicate directly with leaders, organise demonstrations and challenge established media. The same technology also allows false information, manipulated images and fabricated narratives to circulate rapidly, and citizens increasingly inhabit different information environments in which the same event may be interpreted through entirely different narratives. Political disagreement becomes harder to resolve when the factual foundations of the debate are themselves contested.

2. The evidence from public opinion

The Pew Research Center’s survey of 45,658 people in 37 countries, conducted between February and May 2026 and published on October 1, 2026, provides the most comprehensive recent evidence. A median of 68 per cent across the countries said that access to the internet and social media had made people more divided in their political opinions, and a median of 79 per cent said it had made people easier to manipulate with false information and rumours. At the same time, a median of 77 per cent said it had made people more informed about current events, and a median of 60 per cent said it had given people a more meaningful voice in politics. Pew also found that the share of the public regarding social media as bad for democracy has increased significantly in most countries with trend data since 2022 or 2023, although a median of 55 per cent still regarded it as good.

The results are therefore consistent with a dual institutional character: the technology expands participation while increasing exposure to manipulation and fragmentation. The concern is not an American peculiarity. In the United States, 81 per cent said that access to the internet and social media had made people more politically divided, about the same as in 2022, and 64 per cent said social media had been bad for democracy, unchanged from 2022. In Europe, concern has risen sharply: the share saying people had become more politically divided rose in Germany from 65 per cent in 2022 to 82 per cent in 2026, in Poland from 50 to 79 per cent, and in France from 52 to 69 per cent, with 72 per cent in Sweden. In Germany the share saying social media has been bad for democracy rose from 41 per cent in 2022 to 61 per cent. Pew also found that people in wealthier countries are more likely to see social media as harming democracy, with majorities in several lower-income countries remaining positive.

3. Algorithmic amplification and mobilisation

Many platforms organise content through recommendation systems designed to maximise engagement and retention, and content that provokes strong emotion can attract disproportionate attention. In political communication this can create incentives for sensationalism, hostility and simplified explanations of complex economic problems. A reinforcing cycle can result. Economic dissatisfaction creates demand for explanation; political entrepreneurs supply narratives that attribute difficulties to identifiable groups, institutions or foreign powers; algorithmic distribution amplifies emotionally engaging content; repeated exposure reinforces group identity and distrust of other groups; and leaders interpret the resulting mobilisation as evidence of popular legitimacy for more confrontational conduct.

It would nonetheless be incorrect to treat algorithms as independent political actors or to assume that platforms inevitably radicalise users. Their effects depend on platform design, moderation, organised political actors, user behaviour and the surrounding institutional environment, and the same systems also support investigative journalism, civic education and peaceful mobilisation. Pape himself has argued that political violence is not simply an internet phenomenon and that removing individuals from platforms did not end it, which is consistent with the view taken here: information systems amplify and accelerate underlying grievances rather than creating them.

4. Synthetic media and the credibility problem

Generative artificial intelligence lowers the cost of producing convincing text, audio, images and video, and thus the cost of misinformation at scale. Deepfakes and synthetic political content can undermine confidence in authentic evidence, and even when fabricated material is disproved the damage may persist. The danger is an epistemic legitimacy problem: a weakening of society’s capacity to establish a sufficiently shared understanding of events. Elections require citizens to decide on the basis of information, accountability requires a reasonable ability to distinguish evidence from fabrication, and compromise requires at least some common understanding of the problem being debated.

5. Platforms as political infrastructure

Digital platforms are best understood as a form of political infrastructure comparable in systemic importance to traditional institutions of public communication. This does not imply government control of debate. It raises questions of transparency, accountability, political advertising, data protection, algorithmic discrimination and the protection of electoral processes. The United States faces a particularly complicated problem because of the constitutional weight of free expression and the political influence of major technology firms. Europe has pursued a more regulatory path, notably through the Digital Services Act, although regulation cannot by itself resolve political polarisation. The objective should be to protect freedom of expression while improving the transparency and accountability of the systems that distribute political information.

VIII. Institutions and Political Legitimacy: Two Natural Experiments in Autumn 2026

The preceding sections describe pressures. The decisive variable is how institutions respond. Two events in the autumn of 2026, the French budget crisis and the United States midterm elections, offer unusually clear tests of institutional resilience, and their outcomes differ in instructive ways.

1. France: fiscal constraint, parliamentary fragmentation and street mobilisation

Reuters reported on November 5 that France faced its latest nationwide protests after weeks of unrest over the cost of living and public services. The movement began in late September with student protests over underfunded schools and broadened into a wider challenge to the government, including some of the worst street violence and looting in years, notably over the weekend of October 17 and 18. The government mobilised 70,000 police and other security personnel for 220 demonstrations planned for November 5. Authorities put turnout at about 600,000, while unions claimed more than 1.5 million.

The budgetary dimension is central. The government’s proposed 2027 budget includes about 54 billion euros in savings and spending cuts, and on November 4 the National Assembly rejected the revenue section of the finance bill by 458 votes to none, with two abstentions, which under French budget rules sank the text at first reading. The government began releasing 10 million barrels of diesel from strategic reserves on November 1, and the economy minister said pump prices had since fallen by an average of 15 cents a litre. Some unions and housing-rights associations have called for pan-European protests from November 20 to 22. These events occur ahead of the 2027 presidential election, in which the National Rally holds a consistent lead in polling, with its leader Marine Le Pen facing legal uncertainty.

The French case should not be read as a single ideological movement or as proof that European democracy is approaching collapse. Peaceful protest, labour mobilisation and parliamentary opposition are normal components of democratic life, and the unanimous rejection of the revenue measure is as much a symptom of fragmented parliamentary arithmetic as of extremism. The analytical distinction is between legitimate democratic contestation and the normalisation of violence or the rejection of peaceful procedures. What France illustrates is the difficulty of reconciling three competing demands: citizens expect adequate public services and the protection of living standards; markets and European fiscal rules require deficit control; and fragmented legislatures struggle to assemble majorities for either. When these cannot be reconciled, fiscal policy itself becomes a source of political instability, and parties offering simplified alternatives, whether radical redistribution, protectionism, migration restriction or deeper expenditure cuts, stand to benefit.

2. The wider European political landscape

France is not isolated. Polling aggregates in the autumn of 2026 show populist and nationalist parties leading or competing for first place in several of Europe’s largest democracies. In Germany the Alternative for Germany has polled in the high twenties, ahead of the Christian Democratic Union of Chancellor Friedrich Merz, whose government remains unpopular. In Britain, Reform UK led national polls for roughly fifteen months until Labour replaced its leader, and the political position has since been fluid. In France the National Rally leads ahead of the 2027 presidential election. Analysts commonly attribute much of this to anxiety over immigration and the cost of living, although the electorates of these parties differ markedly from one another on foreign policy and on relations with the United States.

An Associated Press analysis published after the United States midterms observed that voter anger has also punished populists who hold office in several countries, which suggests that the underlying driver is generalised anti-incumbent dissatisfaction with living standards and institutional performance, rather than a simple ideological alignment. This reinforces the multidimensional interpretation: the same cost-of-living and institutional-performance pressures can produce opposite electoral results depending on who is in office.

Europe’s institutions also face an accountability gap. Citizens evaluate economic performance at the national level, while many decisions are shaped by European institutions, common regulations and shared fiscal or monetary arrangements. National governments may attribute difficult choices to European constraints, while European institutions point to national implementation failures. The greater the perceived distance between policy decisions and democratic control, the harder it is to sustain confidence. This does not make integration itself the cause of dissatisfaction; it means that multilevel governance requires clear accountability and effective mechanisms for responding to public concerns.

3. The United States: polarisation, political violence and an institutional test passed

The United States presents a different configuration. It combines considerable technological and financial strengths with pronounced inequality, housing pressure and a deep crisis of confidence in public institutions. The central concern in the political-violence literature is the extent to which political violence has become acceptable to parts of the public. Pape’s Chicago Project on Security and Threats reported in an August 2026 national survey that 25 per cent of Americans agreed that force was justified to remove President Trump from office, and found substantial support among some respondents for using the military against protesters. These findings must be read with the other available evidence. Pape himself has said that between 75 and 80 per cent of Americans abhor political violence. The Public Religion Research Institute reported in January 2026 that 77 per cent of Americans disagreed that patriots may have to resort to violence to save the country, against 20 per cent who agreed, and that two-thirds believed that leaders’ failure to condemn violent rhetoric contributes a lot to violence. A separate 2026 study at the University of California, Davis found declining support for political violence on most measures, with roughly three-quarters of Americans rejecting violence for political ends.

The reasonable synthesis is that a large majority rejects political violence while a minority, which amounts to tens of millions of people given the size of the population, regards some forms of force as justifiable under some conditions. Even a minority can threaten institutions when opponents come to be seen as enemies, and the figures measure attitudes, not behaviour. Surveys of reported willingness to endorse violence are an imperfect guide to incidence and must be supplemented by documented incidents, criminal investigations and longitudinal studies.

The November 3 midterm elections offered a test of the system’s capacity to resolve conflict peacefully. The Associated Press and television networks projected that Democrats won control of both the House of Representatives and the Senate, ending the Republican trifecta, with a series of Senate seats flipping from Republican control. President Trump said he was willing to accept the results and to work with Democrats. Amid widespread concern about political violence, the transfer of congressional control through ordinary electoral procedures is evidence that core electoral institutions retained legitimacy. It does not eliminate the risk of later disputes, notably over the 2028 contest, but it is an important data point against deterministic readings of democratic breakdown.

The midterm result also illustrates the interaction with the economic variables discussed above. An election in which living costs, energy prices and affordability featured prominently penalised the governing party in the context of a 3.7 per cent inflation rate in September, energy prices more than 20 per cent higher than a year earlier, and a central bank that has begun to raise rates. This is consistent with the view that economic insecurity translates into political change through normal electoral channels where those channels remain credible.

IX. Comparative Assessment: Different Forms of the Same Crisis

The two regions share structural pressures but differ in how those pressures reach the political system.

1. The American configuration

The American economy has substantial capacity to create industries and attract investment, and the International Monetary Fund projects 2.3 per cent growth in both 2026 and 2027. The benefits of innovation, however, can be highly concentrated, as the record top-end wealth share demonstrates. The absence of sufficiently accessible housing, healthcare and employment protection makes adjustment difficult for households without reserves. The consequences are intensified by a two-party system in which political competition is interpreted through ideological and cultural lines, and by a fiscal and monetary context in which public debt is high and the central bank is tightening.

2. The European configuration

European welfare systems moderate the immediate impact of unemployment and income shocks and the Union’s poverty rate edged down in 2025. But they also face rising fiscal demands as populations age, an energy-exposed industrial base, a growth outlook of just over one per cent, and an institutional structure in which national governments must reconcile domestic priorities with common European commitments. The result is not necessarily greater instability but slower adjustment and dissatisfaction when governments cannot deliver the services and living standards that citizens expect. The French budget crisis shows how quickly such constraints can become acute.

3. The transatlantic technology divide

The United States holds a strong position in foundational artificial intelligence, advanced computing and platform businesses. Europe has important industrial capabilities, research institutions and regulatory influence, but its adoption is uneven. If the United States captures a disproportionate share of artificial intelligence investment and profit, the transatlantic economic relationship could become more asymmetric. Conversely, if European regulation and coordination support trustworthy adoption, workforce development and competitive industrial applications, Europe could retain significant advantages in high-value manufacturing and specialised services. The outcome is uncertain.

4. The common problem: institutional adaptation

The central similarity is that both regions must adapt institutions to economic and technological conditions that are changing faster than traditional political processes. The problem is not simply that citizens are dissatisfied; it is that they increasingly question whether existing institutions can provide economic security, political representation and credible solutions to problems they experience directly.

X. Evaluating the Explanatory Frameworks

The evidence reviewed in this paper allows a comparative evaluation of the leading explanations for democratic strain and political violence, and suggests how each should be extended.

The relative-deprivation and status-threat traditions, including Pape’s work, are supported by the evidence on wealth concentration, housing and generational mobility, and by the persistence of minority support for political violence in surveys. Their limitation is that deprivation and threat are widespread while violence is rare. The transition from grievance to violence requires organisation, elite signalling and triggering events, which the literature on elite norms emphasises.

The economic-geography tradition is strongly supported by the OECD’s 2026 findings and by the Eurostat regional data, which show that grievance is concentrated in particular places and that mobility does not reliably close gaps. Its limitation is that it explains where pressure is greatest but not why the same pressure produces peaceful adjustment in some places and mobilisation in others.

The elite-norms and institutional traditions are supported by the contrast between the French and American cases, and by the observation that polarisation is not incompatible with democracy. A society can be highly polarised yet maintain peaceful competition, whereas a society with moderate ideological differences can experience instability if institutions lose credibility. Their limitation is that they treat structural pressures as background conditions, when those pressures determine the political opportunities available to elites.

The technology-and-power tradition is supported by the evidence on the concentration of artificial intelligence returns and on the distribution of adoption across firms and countries. The evidence so far is ambiguous on aggregate displacement, which counsels humility; the better-supported claim is that technology is altering the distribution of opportunity, particularly at the entry point of careers, and that the political significance lies in perceptions of the future as much as in measured present outcomes.

Taken together, these traditions suggest that the primary dependent variable should not be polarisation as such. It should be the capacity of democratic institutions to maintain legitimate, peaceful and effective political competition under conditions of rapid socioeconomic transformation. Two further extensions follow. First, the distinction between actual and perceived economic displacement should be central: a person may retain employment and rising income yet believe that their position, their children’s prospects and their community’s future are deteriorating because housing is unaffordable, professional status feels insecure or local industry is in decline. Such perceptions should not be dismissed because aggregate indicators are positive, nor treated as proof that objective conditions have worsened for every affected group. Second, the transmission mechanism has changed: digital networks reproduce, personalise and amplify political rhetoric continuously and across borders, so that the interaction of leadership and communications infrastructure is now a first-order variable.

XI. A Bayesian Game-Theoretic Framework

The preceding analysis suggests the need for a framework that integrates structural pressure, institutional mediation and strategic behaviour. This section proposes a conceptual research architecture, not an empirically calibrated forecasting model.

1. The explanatory dimensions

The framework begins with eight interconnected dimensions. Demographic pressure includes ageing, fertility, migration, changes in composition, intergenerational conflict and geographic population shifts. Distributional stress includes wealth concentration, wage stagnation, housing affordability, financial insecurity and unequal access to productive assets. Technological displacement and adaptation includes automation, artificial intelligence adoption, the restructuring of professional employment and the distribution of technological returns. Information-system fragmentation includes algorithmic amplification, misinformation, synthetic media and declining confidence in common sources. Energy and price shock captures the effects of the war involving Iran, the disruption of Hormuz, inflation and the associated monetary response. Governmental and institutional capacity includes fiscal sustainability, administrative competence, public-service delivery and the ability to implement credible reforms. Legitimacy and political trust includes confidence in elections, legislatures, courts, parties and administration. Rhetorical and mobilisation intensity includes elite polarisation, the demonisation of opponents, organised mobilisation and the acceptance of violence as a legitimate instrument.

These dimensions are not independent. Technological displacement can raise inequality if gains accrue to capital owners; inequality increases dissatisfaction; algorithmic systems may amplify narratives that attribute insecurity to specific groups; an energy shock compresses fiscal capacity, which weakens the state’s ability to respond; and weakened responsiveness erodes trust, which raises the payoff to confrontational rhetoric.

2. The causal architecture

The proposed causal architecture has four stages. In the first, structural transformations in demography, distribution, technology, information and energy generate economic insecurity, perceived displacement, declining relative mobility and political distrust. In the second, institutional mediation determines the translation of that pressure: government responsiveness, public services, political leadership, judicial independence and the credibility of electoral processes can absorb it or amplify it. In the third, political entrepreneurs, parties, unions, movements and platforms choose strategies in response to the opportunities that pressure and institutions create. In the fourth, outcomes emerge, ranging from peaceful adjustment and policy reform, through populist mobilisation and democratic erosion, to political violence. These relationships are hypotheses to be tested, not established deterministic effects.

The institutional mediation stage is the critical feature. The same shock produces different outcomes depending on institutional capacity. Automation accompanied by effective retraining, rising productivity and broader distribution of gains may strengthen confidence; the same automation accompanied by job insecurity, concentrated wealth and declining mobility may generate dissatisfaction. The autumn 2026 evidence illustrates the point: comparable cost-of-living pressure led in France to a budget impasse and street unrest, and in the United States to an orderly change in congressional control.

3. The strategic dimension

Because outcomes depend on the choices of actors who anticipate one another’s responses, the framework is game-theoretic as well as Bayesian. Four strategic interactions are especially important.

The first is a fiscal bargaining game among governments, parliamentary blocs, unions, bond markets and European fiscal authorities. In France, each party gains from refusing concessions while others bear the cost of impasse, which resembles a war of attrition, and the unanimous rejection of the revenue measure is a costly signal that no bloc will accept responsibility for austerity. Such games tend to end only when the cost of delay, measured in market stress or public disorder, exceeds the cost of compromise. The second is an elite signalling game in which leaders choose between moderate and confrontational rhetoric. Confrontation pays when audiences reward it and when electoral penalties are low; it becomes costly when voters punish extremity, as the anti-incumbent results in autumn 2026 suggest, and the equilibrium therefore depends on the electoral rules and the cost structure that institutions impose. The third is a game among platforms, regulators and users in which engagement-maximising design competes with transparency obligations, and regulatory credibility determines whether platforms internalise political externalities. The fourth is a coordination problem among central banks, governments and investors, in which a hiking central bank, a government facing protest and a market assessing fiscal sustainability can each worsen the others’ constraints, as the combination of a 4.0 per cent inflation rate, a tightening European Central Bank and a rejected French budget illustrates.

4. Updating on the autumn 2026 evidence

The framework treats historical experience and comparative data as priors and updates them as observations arrive. Several observations since the beginning of autumn warrant a revision of the prior assessments.

For the United States, the peaceful outcome of the midterms, with the president indicating acceptance, lowers the probability of an acute electoral-legitimacy rupture in the near term. The stronger-than-expected October employment data and the resilience of consumption reduce the near-term weight on a labour-market shock. These are offset by a rise in the energy-driven price level, a monetary tightening cycle with a December decision pending, the record concentration of equity wealth, the thinning of the early-career pathway in exposed occupations, and the continued presence of a significant minority endorsing violence in surveys. The net effect is a modest shift in probability toward the managed scenario, accompanied by persistent risk in the unequal-transition scenario.

For Europe, the rejection of the French revenue bill, the scale of the November 5 mobilisation, euro-area inflation of 4.0 per cent, the prospect of a further interest-rate increase into a slowing economy, and the strong polling of populist parties in the largest economies shift probability toward the unequal-transition scenario and raise the tail risk of a compound crisis. Partially offsetting factors are the fall in the Union’s poverty rate, the drawdown of strategic fuel reserves that has already reduced pump prices in France, and the International Monetary Fund’s baseline assumption that Hormuz reopens in March 2027.

5. Three scenarios for 2027 to 2032

The following scenarios are conditional. The probability ranges reported are subjective judgements formed from the evidence reviewed in this paper; they are priors for subsequent updating, not outputs of a calibrated statistical model, and assigning rigorous numerical probabilities would require operational definitions of instability, a calibrated dataset and an estimation procedure.

Scenario A, managed technological and institutional adaptation. Governments invest in digital education, workforce transition, housing and infrastructure, energy security is restored as Hormuz reopens, artificial intelligence adoption broadens beyond large firms, and productivity gains are increasingly reflected in real incomes. Political disagreement continues but institutions retain sufficient legitimacy to manage adjustment. My subjective prior for this pathway is roughly 30 to 40 per cent for the United States and 25 to 35 per cent for Europe.

Scenario B, unequal technological transition and persistent populism. Artificial intelligence adoption accelerates but its benefits remain concentrated; housing affordability remains poor; career entry weakens in exposed professions; energy prices stay elevated for longer than the baseline; and political communication remains fragmented. Populist parties gain by mobilising economic and cultural grievances, and democratic institutions continue to function but with eroding public confidence. My subjective prior is roughly 45 to 55 per cent for both regions.

Scenario C, compound socioeconomic and institutional crisis. Economic stagnation, a financial correction in technology valuations, fiscal stress, a prolonged energy shock, migration tensions and an acute political trigger coincide; elites intensify confrontation, misinformation erodes confidence in institutions, and organised groups challenge the legitimacy of electoral outcomes. Political violence becomes a more serious risk, particularly where conflict-resolution institutions lack credibility. My subjective prior is roughly 10 to 15 per cent for the United States and 15 to 20 per cent for Europe, the higher European figure reflecting thinner fiscal space and the combination of ageing, energy exposure and fragmented legislatures. The probability of widespread political violence is lower than the probability of this scenario, because violence requires additional organisational and triggering conditions.

6. Methodological requirements

A rigorous version of the model must avoid treating the dimensions as statistically independent, since inequality and technological displacement may be correlated and distrust and information fragmentation may reinforce each other; summing individual effects would double-count the same underlying process. It should therefore be implemented as a Bayesian network with conditional dependencies, interaction effects, time lags and country-specific parameters. It must also acknowledge radical uncertainty: historical relationships between productivity, wages and employment may not hold under advanced artificial intelligence, and traditional models of mobilisation may be less reliable when digital communication changes the speed and reach of organisation. A Bayesian approach can update probabilities as evidence arrives but cannot eliminate uncertainty about structural breaks or unprecedented developments, which is why the outputs should be ranges and scenarios rather than point forecasts.

XII. Policy Implications: Reconstructing the Social Contract for a Digital and Energy-Constrained Economy

If the diagnosis presented here is correct, policies directed exclusively toward income redistribution, migration enforcement or political moderation will be insufficient. The objective should be to reconstruct the social contract around the conditions of a digital, automated and energy-constrained economy.

1. Distributing technological productivity

Investment in artificial intelligence, robotics, computing infrastructure and advanced manufacturing should be encouraged, but the distribution of the resulting gains deserves greater attention. Possible instruments include broader employee participation in productivity gains, wider access to pension and investment ownership, competition policy addressing excessive concentration, portable training and transition benefits, worker consultation on the deployment of automation, and incentives that reward investment in workers alongside investment in machines. In Europe, support for small and medium-sized enterprises is critical: the gap between a 55 per cent adoption rate among large firms and 19 per cent among smaller ones will otherwise widen the productivity divide. The objective is to prevent technological progress from creating a permanent division between those who own productive technologies and those who depend exclusively on employment income.

2. Rebuilding the pathway from education to work

The evidence on early-career employment in exposed occupations suggests that education systems and employers must reconsider the transition from study to professional work. Digital literacy, artificial-intelligence-assisted problem-solving, data interpretation, communication and professional judgement should be emphasised, alongside apprenticeships, supervised practice and structured entry programmes that replace the on-the-job learning that automation and remote work are eroding. Because remote work may be as significant as artificial intelligence in this narrowing, employers should also assess how they train and mentor newcomers in distributed arrangements.

3. Place-based and regional policy

Following the OECD’s findings, policy should combine regional development, industrial diversification and employment support rather than relying on labour mobility alone. This includes anticipating skills needs, supporting displaced workers in staying connected to growing sectors and addressing local vulnerabilities before shocks arrive. The Eurostat regional evidence on poverty concentration reinforces the case for targeted rather than purely national instruments.

4. Housing as economic and democratic security

Housing should be treated as part of economic security rather than solely as urban planning. Governments should address constraints on construction, infrastructure finance, land use and the supply of affordable rental and ownership housing. For younger households, access to housing determines the ability to form families, relocate for work and accumulate wealth, and a society in which independence depends on inherited property reinforces intergenerational inequality. The evidence from both regions, a 69 per cent rise in European house prices since 2015 and a 3.2 per cent annual rise in American shelter costs, indicates that supply-side reform is a first-order democratic priority.

5. Managing the energy shock without eroding fiscal space

The International Monetary Fund recommends rebuilding fiscal space, strengthening energy security and improving readiness for artificial intelligence. Temporary and targeted support for vulnerable households, such as the release of strategic reserves in France, is preferable to broad and open-ended subsidies that deepen deficits, and medium-term investment in diversified energy supply reduces exposure to future chokepoint disruptions. Central banks face a genuine dilemma in tightening against supply-driven inflation in slowing economies; clear communication about how they distinguish transitory energy effects from persistent price pressure is essential to preserving credibility without compounding hardship.

6. Migration and integration

Migration policy should reconcile labour-market requirements, demographic sustainability and public confidence in the rule of law. This requires effective border administration, transparent legal pathways, language training, recognition of professional qualifications and investment in local housing, schooling and healthcare. Given that net migration is currently the only source of population growth in the European Union and is projected to be the only source in the United States from 2030, integration capacity is a core economic as well as political variable.

7. Protecting democratic information infrastructure

Governments and democratic institutions should strengthen the transparency and accountability of digital platforms, with particular attention to political advertising transparency, the identification and provenance of synthetic media, independent researcher access to platform data, platform accountability for systemic risks, personal data protection and media literacy. The purpose is not to control political debate but to preserve the conditions under which citizens can distinguish authentic evidence from deliberate manipulation and participate meaningfully in decisions.

8. Reforming public institutions and norms

The effectiveness of democratic institutions depends on their capacity to deliver tangible results. Governments must become more responsive to housing shortages, healthcare pressures, employment transitions and regional disparities, through improved administration, transparent fiscal priorities and credible mechanisms for evaluating performance. Leaders across the political spectrum bear responsibility for avoiding the treatment of every disagreement as an existential conflict, since survey evidence indicates that Americans of both parties regard leaders’ failure to condemn violent rhetoric as a major contributor to political violence. The functioning of the federal and European budget processes, as the French crisis shows, also deserves attention: institutional rules that allow every actor to veto and none to govern are themselves a source of legitimacy loss.

9. A transatlantic strategy for technological and socioeconomic resilience

Europe and the United States could cooperate on artificial intelligence safety and productivity research, workforce transition programmes, competitive digital infrastructure, responsible data governance and industrial innovation. Such cooperation is particularly important if technological concentration becomes a source of economic dependence or strategic vulnerability, and it would strengthen both regions’ capacity to treat the digital transformation as a means of reinforcing rather than weakening the social foundations of democratic capitalism.

XIII. Conclusion: Institutional Adaptation in the Age of Artificial Intelligence

The socioeconomic challenges confronting Europe and the United States are more extensive than any single explanation of populism, demographic anxiety or inequality suggests. Demographic transformation, concentrated wealth, a structural shift in the technologies of production, the reorganisation of public information, an energy and price shock and the weakening capacity of public institutions are interacting. Each has been identified by an important body of scholarship, including work on relative deprivation, economic geography, elite norms and the political economy of technology; Robert Pape’s work on violent populism is a significant contribution to the first of these. The argument developed here is that they should be treated as a system.

That system has a feedback structure. An ageing population requires productivity growth, which encourages automation; automation may weaken employment security for some workers; inequality concentrates ownership of the technologies; digital communication amplifies the resulting dissatisfaction; an energy shock reduces the fiscal room to respond; political entrepreneurs exploit grievance; and institutional weakness makes credible response more difficult.

1. The fundamental issue is institutional adaptation

The most important question facing mature democracies is not whether demographic, economic and technological transformations can be prevented. They cannot. It is whether institutions can adapt quickly enough to manage their consequences. A democratic system must demonstrate that it can provide opportunity, protect individual rights, maintain social mobility and distribute the benefits of technological progress reasonably fairly. If it fails, dissatisfaction may become detached from party competition and institutional processes. This does not make political violence inevitable, but it creates conditions in which entrepreneurs of resentment find it easier to undermine confidence in democratic procedures.

2. Technological citizenship

In the industrial economy, employment, wages, collective bargaining and social insurance formed the foundations of the relationship between citizens, employers and the state. In the emerging economy these arrangements may no longer suffice. Citizens will need access to digital education, opportunities for continuous professional development, protection during technological transitions and a reasonable opportunity to share in the gains generated by artificial intelligence. The ownership and governance of productive technologies will become central questions of political economy. The aim is not to prevent automation or preserve every existing occupation but to ensure that productivity gains support social mobility, economic security and broader prosperity.

3. A new measure of democratic resilience

Resilience should be measured not only by the survival of elections and constitutional institutions but by whether citizens believe that the system offers a credible future. A democracy in which elections continue but younger generations cannot achieve economic independence, workers cannot adapt to technological change and institutions cannot respond to social pressure may remain formally democratic while experiencing a gradual loss of legitimacy. Conversely, a society can undergo substantial transformation without instability if its institutions remain responsive and its citizens retain confidence in peaceful competition. The autumn 2026 evidence from France and the United States illustrates both possibilities.

4. Final proposition

The contemporary crisis of mature democracies is best understood as an institutional adaptation crisis arising from the simultaneous transformation of demographic structures, wealth distribution, productive technologies, information systems and the global energy and price environment. Political violence and populist mobilisation are possible consequences of that process, not its only form. The future of European and American democracy will depend on whether governments can reconcile technological productivity with social inclusion, demographic change with fiscal sustainability, and political pluralism with institutional legitimacy. The decisive issue is not whether society will change but whether democratic institutions can manage that change without sacrificing the economic security, political equality and public trust on which their legitimacy depends. The central challenge for 2027 to 2032 is to construct an inclusive democratic political economy capable of governing the digital and artificial intelligence revolution while preserving the social foundations of democratic capitalism.

XIV. Research Limitations and Methodological Qualifications

Several qualifications are important. First, the relationship between demographic change and political violence is not deterministic; outcomes depend heavily on institutional capacity and policy. Second, artificial intelligence exposure should not be confused with actual displacement: a technology’s capacity to perform a task does not establish that the occupation containing it will disappear, and the timing and scale of employment effects remain uncertain, as the disagreement between analyses emphasising artificial intelligence and those emphasising remote work shows. Third, wealth concentration, technological adoption and political polarisation are interconnected, so their statistical relationships must be analysed to avoid double-counting. Fourth, opinion surveys on political violence measure attitudes and reported willingness, not behaviour, and different surveys with different question wording produce different magnitudes. Fifth, several data series are affected by the 2025 federal government shutdown, which prevented the collection of some October 2025 data and therefore the publication of certain year-on-year comparisons for October 2026. Sixth, Eurostat figures cited for 2025 reflect the reference year of the relevant survey, and some incomes refer to 2024. Seventh, the scenario probabilities in this paper are subjective priors and the Bayesian framework is a conceptual architecture whose next stage requires comparable datasets for the United States, major European economies and the European Union.

The analysis uses information published on or before November 10, 2026. Where complete November data were not available, the latest published observations are identified by their actual reference periods. Evolving events, including the French budget process, the December meetings of the Federal Reserve and the European Central Bank, and developments in the Strait of Hormuz, may alter the assessment.

Selected References and Documentary Sources

A. Scholarly and book-length works

Acemoglu, D. (2025). The simple macroeconomics of AI. Economic Policy.

Acemoglu, D., and Johnson, S. (2023). Power and Progress: Our Thousand-Year Struggle over Technology and Prosperity. New York: PublicAffairs.

Acemoglu, D., and Robinson, J. A. (2019). The Narrow Corridor: States, Societies, and the Fate of Liberty. New York: Penguin Press.

Autor, D. H., Dorn, D., and Hanson, G. H. (2013). The China syndrome: Local labor market effects of import competition in the United States. American Economic Review, 103(6).

Brynjolfsson, E., Chandar, B., and Chen, R. (2026, revised August). Canaries in the Coal Mine? Six Facts About the Recent Employment Effects of Artificial Intelligence. Stanford Digital Economy Lab.

Case, A., and Deaton, A. (2020). Deaths of Despair and the Future of Capitalism. Princeton: Princeton University Press.

Gurr, T. R. (1970). Why Men Rebel. Princeton: Princeton University Press.

Levitsky, S., and Ziblatt, D. (2018). How Democracies Die. New York: Crown.

Norris, P., and Inglehart, R. (2019). Cultural Backlash: Trump, Brexit, and Authoritarian Populism. Cambridge: Cambridge University Press.

Pape, R. A. (2026). Our Own Worst Enemies: America in the Age of Violent Populism.

Piketty, T. (2014). Capital in the Twenty-First Century. Cambridge, MA: Harvard University Press.

Rodrik, D. (2018). Populism and the economics of globalization. Journal of International Business Policy, 1.

Walter, B. F. (2022). How Civil Wars Start: And How to Stop Them. New York: Crown.

B. Official statistics and intergovernmental institutions

Board of Governors of the Federal Reserve System. Distributional Financial Accounts, release of September 18, 2026 (second quarter 2026).

Board of Governors of the Federal Reserve System. Financial Accounts of the United States, Z.1, release of September 11, 2026.

Board of Governors of the Federal Reserve System. FOMC statement and Chair’s press conference, October 28, 2026.

U.S. Bureau of Labor Statistics. The Employment Situation, October 2026, November 6, 2026.

U.S. Bureau of Labor Statistics. Consumer Price Index, September 2026, October 14, 2026; and October 2026 release of November 10, 2026.

Congressional Budget Office. The Demographic Outlook: 2026 to 2056, January 7, 2026.

Eurostat. EU Population Continued to Grow in 2025, July 10, 2026.

Eurostat. People at Risk of Poverty or Social Exclusion, 2025 data, April 30, 2026; and Regional Divide in Risk of Poverty or Social Exclusion, October 2026.

Eurostat. Digitalisation in Europe, 2026 Edition; and The Use of Artificial Intelligence Technologies in the European Union: Key Results, 2026 Edition.

Eurostat. Housing cost overburden statistics, 2025 data, and Housing in Europe dashboard, 2026; and euro area flash estimate of consumer prices, November 4, 2026.

European Central Bank. Monetary policy decisions of September 10 and October 29, 2026.

International Monetary Fund. World Economic Outlook: Resilience Under Strain, Urgent Choices, October 13, 2026.

Organisation for Economic Co-operation and Development. OECD Employment Outlook 2026: Geographic Disparities in Jobs and Incomes, July 2026.

Organisation for Economic Co-operation and Development. AI Adoption by Small and Medium-Sized Enterprises, report prepared for the G7 presidency, December 2025.

Federal Reserve Bank of New York. Labor market for recent college graduates, quarterly tracker (second quarter 2026), and June 2026 analysis of remote work and young-graduate unemployment.

U.S. Census Bureau. Decennial census data on race and ethnicity, 1990 and 2020, and subsequent population estimates.

C. Surveys and polling research

Pew Research Center. Across the Globe, People Increasingly Say Social Media Is Harming Democracy, October 1, 2026 (Spring 2026 Global Attitudes Survey, 37 countries).

Pew Research Center. Americans Are Negative About Social Media’s Impact on Democracy, and the Rest of the World Is Catching Up, October 2026.

Public Religion Research Institute. Political Violence in America: Public Perceptions, Polarization, and Accountability, January 16, 2026.

Fairlie, R. W., and Wu, J. (2026). Early impacts of artificial intelligence on employment among recent college graduates, analysis of Current Population Survey data for June to August 2026.

D. Reporting by wire services and established news organisations

Reuters. Explainer: France Protests: What Is Behind the Unrest and What Comes Next?, November 5, 2026.

Euronews. France braces for nationwide strike as government backs austerity budget; and France hit by nationwide strike after MPs throw out austerity budget, November 5, 2026.

Associated Press. Election 2026: Democrats capture both chambers of Congress, November 4, 2026; and analysis of the populist right after the midterms, November 7, 2026.

CBS News. Inflation remained elevated in October, CPI report shows, November 10, 2026.

Irish Times. US midterm elections: Democrats take control of House and Senate, November 4, 2026.

Texas Public Radio. The Source: Growing acceptance of political violence raises alarms about American democracy (interview with Robert Pape), October 27, 2026.

PolitiFact. What polls show about political violence in the United States, May 4, 2026.

NPR. A look at research on Americans’ changing attitudes toward political violence (interview with Robert Pape), September 12, 2025.

Farid Novin

Global Review, November 10, 2026