Inequality & Social Cohesion
Very high income and wealth inequality; the top one per cent holds more equities than the bottom 90 per cent combined.
Where does the USA stand, and where does it break?
Wealth inequality in the United States ranks among the highest of the major industrialised nations. In 2025, the gap reached its widest level in decades, according to the Federal Reserve. The top one per cent of households holds roughly one third of total net wealth, whilst the bottom half holds only a few per cent (Federal Reserve, Distributional Financial Accounts). Equity ownership is particularly unevenly distributed: the top one per cent holds more equity wealth than the bottom 90 per cent combined, and the wealthiest ten per cent own the vast majority of all shares. This is precisely the form of wealth that is growing most rapidly in the AI economy. Income inequality, as measured by the Gini coefficient, stands well above the OECD average. At the same time, trust in government, media and institutions is low (Edelman Trust Barometer), and political polarisation between the two camps runs deep (Pew Research). Social mobility — once the cornerstone of the American promise — is distributed unevenly both geographically and across generations. High student debt and unequal schooling constrain mobility. Immigration keeps society young, but also sharpens cultural fault lines.
AI raises productivity; the returns accrue as capital gains. Where capital is unevenly distributed, inequality grows. The United States is particularly exposed here: because equity ownership is extremely concentrated, the gains from automation flow predominantly to those who are already wealthy. The middle class — whose white-collar and knowledge work AI is transforming — faces a dual squeeze: on income and on access to capital returns. When this economic divide meets low institutional trust and deep polarisation, the social fabric frays. Federalism also allows states to drift in very different directions, making common responses harder to achieve. Will the United States distribute the gains from AI more broadly and strengthen social cohesion, or will the divide become entrenched? This is the pivotal social question on the American path.
Paths for the USA
The United States opens access to the capital gains of the AI economy to broad sections of the population. This can take the form of automatic retirement saving, subsidised savings accounts or citizen-fund approaches that make lower-income households co-owners as well. This strikes at the heart of the fault line, since the gains accrue to concentrated capital. The high share-ownership rate of around 62 per cent provides a foundation on which to build. The limit: those on low incomes can barely save. Without substantial subsidies, the impact for the bottom half remains limited.
Trade-off: participation versus the fiscal cost of subsidies
Feasibility: Effective against the divide only with genuine subsidies
The United States strengthens redistribution through progressive taxation of high incomes, capital gains and wealth, as well as through transfers such as the Earned Income Tax Credit. This matters because it addresses concentration directly, where capital participation alone falls short. Existing instruments such as the EITC have demonstrably reduced poverty. The limit lies in political deadlock and capital mobility: higher capital and wealth taxes are contested and risk prompting evasion and capital flight among the wealthy.
Trade-off: distributional equity versus capital flight and locational competition
Feasibility: Traditionally difficult to win majority support for in US politics
The United States secures social mobility through affordable education, retraining and the removal of access barriers, so that background does not determine whether people can participate in the new economy. This matters because mobility is the cornerstone of the American promise, and high student debt constrains it. Community colleges and reskilling programmes offer starting points. The limit: in the United States, education is strongly tied to local resources and income, so that unequal schools continue to hamper mobility.
Trade-off: equal opportunity versus structural educational inequality
Feasibility: Works over the long term, but slowly against entrenched patterns
The United States invests purposefully in social cohesion, local community and the restoration of institutional trust in order to counter polarisation. This matters because economic division and low trust reinforce one another. Local initiatives and cross-party formats offer starting points. The limit lies in the depth of polarisation and in federalism: trust cannot be mandated, and partisan division makes it harder to mount precisely those collective efforts that would strengthen the social fabric.
Trade-off: cohesion versus the reality of polarisation
Feasibility: Requires local roots and time, not programmes alone
Trigger-based phases
Trigger: AI gains flow predominantly to concentrated capital, whilst the middle class faces pressure on both income and access to capital returns.
The wealth gap is widening and social mobility is stalling. Economic divisions are compounding low institutional trust and deep polarisation. The question of distribution is moving to the centre of politics.
Trigger: Growing inequality and eroding social cohesion are increasing pressure for collective responses — responses that federalism makes difficult to deliver.
Debates around participation, redistribution and education are coming to the fore as states drift apart. Whether this pressure is channelled constructively into participation models or destructively into further division remains an open question.
Trigger: The decisive question is whether the gains from AI are distributed more broadly and social cohesion strengthened, or whether the divide becomes entrenched.
If broad participation and a minimum of shared trust can be achieved, society will hold together through the upheaval. If the divide hardens, social peace will come under strain. This fork in the road will shape the cohesion of the United States through to 2040.
Real-world pilot programmes
Status: In force (largest income-boosting transfer programme for working individuals)
The EITC supplements low earned incomes through the tax system and has measurably reduced poverty. As an established instrument with cross-party support, it is a test case for whether participation in prosperity can be strengthened from the labour side — as a complement to the capital side.
Status: Being implemented (state-funded wealth accounts for newborns)
Connecticut opens a state-funded wealth account for children from low-income families. As a federal laboratory, the programme tests how wealth accumulation can be broadened from birth — a direct approach to countering the concentration of capital.
Status: Being piloted (dozens of municipal pilot programmes with direct payments)
Numerous US cities, spurred in part by the Mayors for a Guaranteed Income network, are piloting regular direct payments to low-income households. They are testing whether a partial income floor can cushion transitions where the national safety net is thin.
What we are watching
This measures the degree to which wealth is concentrated at the top. The indicator matters because a further rise in the share of AI gains flowing to the top would entrench the divide. If that share increases, the growth in prosperity accrues to a narrow stratum. For social cohesion, this is the central distribution indicator.
Trust measures the strength of the social fabric. The indicator matters because declining trust and rising inequality reinforce one another. If trust falls further, collective action becomes harder. For the stability of democracy during upheaval, it is a critical variable.
Mobility measures whether background determines life chances. The indicator matters because declining opportunity hollows out the American promise and entrenches the divide. If it falls, class lines harden. For cohesion and legitimacy, it is a core indicator.