Illness, an accident, job loss, the birth of a child, an inability to work, or simply old age. Behind each of these events lies the same question: what happens to an individual when income disappears while basic needs remain?

For centuries, societies answered this question through family support, community solidarity, charity, or personal savings. Gradually, however, another architecture emerged: social protection. Mandatory contributions, public insurance, pensions, benefits, compensation schemes, and solidarity mechanisms transformed risks that had once been largely private into responsibilities shared, at least partly, by society as a whole.

This architecture has become so deeply embedded in developed economies that it can appear almost natural. It is not. It represents one of the largest redistribution systems ever created and one of the central pillars of the modern social contract.

But that contract is coming under increasing pressure. Populations are ageing. Careers are becoming less linear. Informal employment remains widespread across much of the world. Family structures are changing. Healthcare costs are rising. And governments must finance substantial social commitments while the growth of their working-age populations is slowing, and in some cases reversing.

The question is therefore no longer simply how to protect individuals. It is how to preserve that protection when the society for which it was designed is changing structure.

Turning Individual Uncertainty into Collective Risk

The fundamental principle of social protection is risk pooling.

An individual does not know whether they will become seriously ill, lose their job, or suffer a disability that prevents them from working. Across an entire population, however, these risks become considerably more predictable. By pooling contributions and risks, societies can transform a succession of individual uncertainties into collective mechanisms that are relatively manageable.

The logic resembles insurance, but its ambition extends further. Social protection does not merely compensate for a loss. It also seeks to guarantee a degree of economic and social continuity.

Income earned through work therefore becomes the foundation of a system in which part of the wealth generated during periods of activity finances periods when work becomes impossible, interrupted, or permanently ends.

This architecture can cover illness, maternity, workplace accidents, unemployment, disability, old age, long-term care, and family responsibilities. Depending on the country, these protections are financed by workers, employers, taxation, funded schemes, or some combination of all four.

There is therefore no single global system of social protection, but rather a multitude of national compromises between individual responsibility, occupational solidarity, and state intervention.

Bismarck, Beveridge and Two Great Traditions

Two historical models continue to shape much of today's social protection architecture.

The first is associated with the social reforms introduced under Otto von Bismarck in late nineteenth-century Germany. Protection is largely connected to employment and contributions. Workers and employers finance insurance schemes that create entitlements to benefits. The underlying logic is contributory: workers participate in the system during their working lives and, in return, acquire protection against certain risks.

The second model is associated with the report published in 1942 by British economist William Beveridge. Its logic is more universal. Social protection is no longer attached exclusively to occupational status; it becomes part of citizenship and a mechanism for combating poverty. Financing relies more heavily on taxation, while benefits are available to a broader share of the population.

In practice, these two traditions have gradually converged.

France, Germany, and Belgium retain a strong insurance-based and contributory dimension. Nordic countries have developed more universal mechanisms financed extensively through taxation. The United Kingdom combines public benefits with occupational arrangements. The United States operates major federal programmes such as Social Security and Medicare while giving considerable importance to private insurance and employer-sponsored benefits.

No major system is entirely public or entirely private. Modern social protection systems are almost invariably hybrids.

Pensions at the Financial Core

Among all social protection mechanisms, pensions occupy a particular position because they organise an income transfer that can last for decades.

There are two principal ways of financing them.

Under a pay-as-you-go system, contributions collected today directly finance pensions paid today. The arrangement therefore depends on an implicit contract between generations: workers finance current retirees on the assumption that future generations will do the same when they themselves leave the labour market.

Under a funded system, contributions are invested in financial assets to accumulate capital that will eventually finance the contributor's retirement income.

Neither mechanism eliminates risk. They merely distribute it differently.

Pay-as-you-go systems depend heavily on demographics, employment, wages, and the political capacity to maintain sufficient contributions. Funded systems depend more directly on investment returns, inflation, interest rates, longevity, and financial-market stability.

Many countries consequently combine several pillars: public pensions, occupational schemes, and individual savings.

This diversification also explains the growing importance of pension funds in global finance. Savings accumulated to finance old age are no longer simply a social-policy issue. They constitute an enormous pool of capital invested across bonds, equities, real estate, and infrastructure.

The Demographic Problem

For much of the twentieth century, pension systems in developed economies benefited from a favourable demographic configuration: relatively young populations, expanding employment, rising wages, and large numbers of contributors supporting a smaller population of retirees.

That configuration is gradually disappearing.

According to OECD projections, the number of people aged 65 and over per 100 people aged 20 to 64 across its member countries is expected to rise from around 33 in 2025 to 52 by 2050.

This shift fundamentally changes the equation.

If the retired population grows faster than the workforce, four broad options emerge: increase contributions, reduce pensions relative to previous income, raise retirement ages, or allocate more government revenue to the system.

None is politically painless.

This is why debates over retirement age are often presented as purely social conflicts when they also reflect an arithmetic constraint. How the cost of ageing is distributed becomes a political choice: how much should be borne by workers, retirees, employers, taxpayers, or future generations through public debt?

A Deeply Unequal Global Landscape

The paradox of social protection is that it has become almost invisible where it is most developed, while remaining inaccessible to a vast part of humanity.

The International Labour Organization estimates that 52.4% of the world's population is now covered by at least one social protection benefit. It is a historic threshold: for the first time, a majority of humanity has some form of coverage. Yet it also means that 47.6% of the global population — approximately 3.8 billion people — still lacks any effective social protection.

The differences between countries are immense.

High-income economies generally possess long-established institutions, administrations capable of collecting contributions and taxes, and labour markets that are largely formalised. In economies where informal employment dominates, the situation is radically different.

A contributory system is difficult to sustain when a large proportion of the population works without formal contracts, regular income declarations, or consistent participation in administrative systems.

This challenge is particularly significant across parts of Africa, Asia, and Latin America. The central issue there is not simply making existing systems more generous. It is expanding their reach.

Economic formalisation therefore becomes a social question as much as a fiscal one.

The Cost of Solidarity

Social protection also requires the mobilisation of substantial economic resources.

According to the ILO, countries spent an average of 12.9% of GDP on social protection excluding healthcare in 2023, with another 6.5% allocated to health. These global averages naturally conceal enormous differences depending on levels of development and national institutional structures.

Yet these expenditures cannot simply be understood as costs.

Unemployment insurance prevents the loss of a job from immediately becoming a financial catastrophe. Pensions reduce poverty among older people. Health insurance limits the risk that illness will destroy a household's wealth. Family benefits can mitigate the economic cost of raising children.

Social protection therefore acts as a stabiliser.

During recessions, income distributed through social mechanisms supports part of household consumption precisely when market incomes are contracting. Redistribution consequently performs a macroeconomic function in addition to its social one.

But this stabilising capacity comes with a requirement: governments and institutions must be able to collect, administer, and redistribute resources sustainably.

How Much Income Should Be Replaced?

The generosity of a social protection system cannot be measured simply by whether a benefit exists. What matters is how much income it actually preserves.

For pensions, this is usually expressed through the replacement rate: retirement income relative to earnings during working life.

Across OECD countries, the future gross replacement rate from mandatory pension schemes averages around 52% for a worker earning the average wage over a full career. After accounting for taxes and contributions, the average future net replacement rate is around 63%. But differences between countries remain substantial.

These variations reflect fundamentally different conceptions of the state's role.

Some systems seek primarily to prevent poverty, leaving individuals and employers largely responsible for maintaining previous living standards. Others seek to preserve a much larger share of pre-retirement income.

The choice is never purely technical. It reflects a political conception of solidarity.

Work Itself Is Changing

Another difficulty is now emerging: many social protection systems were built around an economic figure that is becoming less universal — the employee working for several decades through a relatively stable succession of formally declared jobs.

Modern careers can be far more fragmented.

Self-employment, digital platforms, temporary contracts, periods of unemployment, international mobility, and frequent changes in employment status complicate the traditional relationship between work and social protection.

When an entitlement depends on the number of years of contributions, any interruption in employment can have consequences that become visible only decades later.

The issue is particularly important for self-employed workers and for people whose careers include long periods devoted to unpaid activities, especially caring for children or dependent relatives.

A system designed around twentieth-century industrial employment must gradually learn how to protect the fragmented working lives of the twenty-first century.

Funded Pensions: Finance in the Service of Protection?

As populations age, funded pension systems are regularly presented as an alternative to pay-as-you-go arrangements.

They do offer one important advantage: part of future retirement income depends on accumulated capital rather than directly on the immediate ratio between workers and retirees.

But this introduces another dependency.

Future living standards become connected to financial-market performance. A generation retiring after several decades of strong market returns may accumulate considerable wealth. Another exposed to prolonged weak returns, a financial crisis, or high inflation may face a very different outcome.

Funded systems also raise a distributional question. Higher-income households can generally save more and therefore benefit more extensively from capital growth. Without corrective mechanisms, retirement systems can reproduce, or even amplify, inequalities accumulated during working life.

This is why resilient systems increasingly seek not to choose permanently between pay-as-you-go and funded models, but to distribute risks across several mechanisms.

The OECD estimates that private pensions now play a significant role in more than one-third of its member countries. For an average-income worker completing a full career, the projected gross replacement rate rises from around 43% from public schemes alone to approximately 52% when all mandatory arrangements are included.

The Real Contract Between Generations

Behind the technical parameters of social protection lies a much deeper question.

Who should bear the risks of existence?

The individual? The family? The employer? The state? Future generations? Financial markets?

Every social system offers a different answer, but none can abolish the cost of risk. It can only decide how that cost will be distributed.

A generous pension must be financed. Universal coverage must be administered. Unemployment insurance requires contributions when employment is abundant. Health insurance mutualises expenditure that exists regardless of how it is financed.

Social protection is therefore never free. But neither is its absence.

A society that substantially reduces collective risk-sharing simply transfers more risk to households. Precautionary savings, family support, debt, loss of wealth, or poverty then become the mechanisms through which adjustment occurs.

A System to Reinvent, Not Simply Dismantle

Debates over the future of social protection are often trapped between defending existing systems and reducing public expenditure.

The real challenge is more complex.

Social protection systems must remain generous enough to fulfil their purpose, broad enough to cover emerging forms of work, and financially resilient enough to withstand demographic ageing.

This will probably require increasingly hybrid architectures: universal foundations financed through taxation, contributory insurance, occupational pensions, complementary funded schemes, specific arrangements for self-employed workers, and social rights attached more closely to individuals rather than to a particular employer.

In countries where informal employment remains dominant, the priority will be different: gradually building protection capable of extending beyond formal salaried employment.

The history of social protection is the history of a fundamental transformation. Events once regarded primarily as private misfortunes became risks that society agreed, at least partially, to share.

That transformation profoundly altered the relationship between individuals, work, and the state.

It is now entering a new phase.

Because behind the debates over contributions, pensions, deficits, benefits, and retirement ages lies a much simpler question: how much of the risk of being alive is a society willing to carry collectively?

Main Sources

International Labour Organization — World Social Protection Report 2024–26: Universal Social Protection for Climate Action and a Just Transition

OECD — Pensions at a Glance 2025: OECD and G20 Indicators

OECD — research on public and private pension systems and pension replacement rates