The Mixed Economy: Welfare States, Keynes, Neoliberalism and State Capitalism

Follows: Where Public Money Goes · The Political Spectrum, Fairly Presented · Capitalism: How Markets, Prices and Private Property Organise an Economy

In this deep dive

1. Bismarck's bargain: the conservative invention of social insurance
2. The Depression and the Keynesian revolution
3. Building the floor: the New Deal and the Beveridge Report
4. The golden age, and the three worlds of welfare capitalism
5. Stagflation and the turn of the 1980s
6. "Neoliberalism": a word worth handling with care
7. 2008 and after: the state comes back, again
8. State capitalism: China, Singapore and the sovereign wealth funds
9. Social spending compared, and what the numbers can't tell you

In 1889 a German factory worker who reached seventy, and who had paid into a new state scheme for thirty years, became entitled to a pension. Very few lived that long. The law was modest, almost symbolic. It was also the start of something that now takes about a fifth of the national income of the average rich democracy and shapes nearly every life in it, from the hospital you were born in to the pension you will draw.

The earlier pieces in this series looked at economic systems as pure types. Capitalism showed how prices and private property coordinate production; Socialism: The Many Kinds, From Utopians to Social Democracy and Communism: The Idea, the States, and What Happened look at the case for collective ownership and planning. In practice, every rich country chose neither. They built mixed economies: mostly private ownership and markets, plus a state that insures people against the big risks of life, manages the business cycle, regulates, and in some places owns or steers large parts of industry.

Where Public Money Goes (Politics 8) explained the arithmetic of a public budget, and The Great Political Traditions (Politics 15) explained social democracy, Christian democracy and classical liberalism as political families. This piece is about the mixture itself: where each ingredient came from, why the recipe kept changing, and what the evidence says. On the values underneath it takes no side; advocates of a bigger state and of a smaller one can both point to prosperous countries that chose their way.

Part 1 — Bismarck's bargain: the conservative invention of social insurance

The welfare state was not invented by socialists. It was invented by one of the most conservative statesmen of the nineteenth century, partly to stop socialists.

Germany in the 1870s and 1880s was industrialising fast. Millions of people had moved from farms, where family and village absorbed sickness and old age, to factories, where an injury or a fever could mean destitution. A socialist workers' movement was growing. Otto von Bismarck, the imperial chancellor, responded with three laws in six years:

The German pension insurer, the Deutsche Rentenversicherung, describes the motive plainly in its own history: Bismarck "recognised the danger" (our translation) and used these laws to bind the working class more closely to the state and prevent its radicalisation. The pension contribution was about 2% of pay, split equally between employer and worker, with a small subsidy from the state. By 1891 about 126,400 old-age pensions were being paid, thanks to a transition rule for people already over seventy.

The first welfare state was a conservative project: a way to give workers a stake in the existing order so they would not try to overturn it.

Three features of Bismarck's design still matter, because whole countries still follow them.

It was insurance, not charity. Benefits were earned by contributions, tied to your job and your wage. That made them respectable: you were collecting what you had paid for, not begging. It also made them unequal: people outside formal paid work, which then included most women, were left out or depended on a working relative.

It was run by funds, not ministries. Employers and workers paid in and had a say in running the insurance bodies, and much of continental Europe still organises social insurance through funds of this kind.

It was a bargain. The state offered security in return for loyalty. Supporters of social insurance across the political spectrum have made versions of this argument ever since: a market economy is more stable, and its people freer, when no one fears that one illness will ruin them. Critics on the left saw it as a bribe to buy off demands for deeper change. Critics on the right saw the first step towards a state that would take over responsibilities families and churches had always carried. Both criticisms are still heard today.

The idea spread. Over the next half-century, versions of social insurance appeared across Europe and beyond. But in 1930 public social spending was still small almost everywhere. On the long-run series compiled by the economic historian Peter Lindert and the OECD, it was about 2% of GDP in Britain, under 5% in Germany and well under 1% in the United States. Then came the Depression.

Part 2 — The Depression and the Keynesian revolution

The crash came on "Black Thursday", 24 October 1929. What followed was not a recession but a collapse. By the time Franklin Roosevelt was inaugurated on 4 March 1933, the American banking system had broken down and, according to the Roosevelt Presidential Library, 24.9% of the workforce, nearly 13 million people, was unemployed. Similar disasters hit Germany, Britain and most of the industrial world.

The economics of the time had a hard time explaining this. The standard view held that markets clear: if workers are unemployed, wages fall until employers want to hire them; if savings pile up, interest rates fall until someone wants to borrow and invest. Unemployment on this scale should be temporary. It wasn't.

What Keynes argued

John Maynard Keynes (1883–1946), a Cambridge economist, published The General Theory of Employment, Interest and Money in 1936. Its central claims, stripped to their bones:

1. Spending creates income. One person's spending is another person's wages. Total demand in the economy (consumers, businesses investing, government, buyers abroad) determines how much gets produced and how many people get hired.
2. Demand can stay too low for years. If households and firms are frightened, they save more and invest less. Each person's caution is sensible; added together, it shrinks everyone's income. There is no automatic force that quickly brings the economy back to full employment.
3. Cutting wages may not help. A general wage cut, which the old view recommended, also cuts workers' incomes and therefore their spending, which can make demand weaker still.
4. So the state should fill the gap. In a slump, government can borrow and spend, or cut taxes, to replace the missing private demand. Monetary policy helps but may not be enough on its own.

Keynes also had a striking account of why investment is so unstable. Business decisions about the far future, he wrote in chapter 12, cannot be the outcome of careful probability sums. They are taken "as a result of animal spirits — of a spontaneous urge to action rather than inaction." When animal spirits fail, investment collapses, and with it jobs.

"The outstanding faults of the economic society in which we live are its failure to provide for full employment and its arbitrary and inequitable distribution of wealth and incomes." — Keynes, The General Theory, ch. 24

What Keynes did not argue

It is easy to make Keynes more radical than he was. He wanted, in his words, "a somewhat comprehensive socialisation of investment", meaning the state would take responsibility for keeping total investment high enough. But he was explicit that once full employment was secured, there was "no more reason to socialise economic life than there was before" (ch. 24). He thought markets did a good job of deciding what to produce and how; his complaint was that they did not reliably produce enough. Keynes saw himself as saving capitalism from its own worst failure, not replacing it.

The case against, at its strongest

Keynes's critics were formidable, and some of their objections have aged well.

• Friedrich Hayek argued that slumps were the hangover from earlier booms fuelled by cheap credit, and that government spending would delay the necessary readjustment of production rather than cure it.
• Milton Friedman and Anna Schwartz later argued that the Great Depression was made far worse by the Federal Reserve letting the money supply collapse. On that reading the lesson was "get monetary policy right", not "expand the state".
• Public-choice economists worried about politics: governments find it easy to run deficits in bad times but hard to run surpluses in good ones, so Keynesian policy would ratchet the state and its debts ever upward.

The modern consensus, to the extent there is one, borrows from both sides: central banks fight ordinary downturns, and governments use the budget in deep slumps, especially when interest rates are already near zero. Keynes himself predicted his ideas would win slowly. "Practical men, who believe themselves to be quite exempt from any intellectual influences," he wrote, "are usually the slaves of some defunct economist."

Part 3 — Building the floor: the New Deal and the Beveridge Report

The New Deal

Roosevelt had pledged "a New Deal for the American people". Much of it came before Keynes's book and owed more to improvisation than to theory. In his first years in office came the Federal Deposit Insurance Corporation (1933), which guaranteed bank deposits and ended the runs; the Tennessee Valley Authority (1933), a public corporation building dams and electrifying a poor region; the Securities and Exchange Commission (1934); and the Works Progress Administration (1935), which hired the unemployed for public works.

The most lasting piece was the Social Security Act of 14 August 1935. It created federal old-age benefits, unemployment insurance run with the states, and aid for dependent children and for blind and disabled people. Like Bismarck's system, the pension was contributory, financed by payroll taxes and tied to work.

The New Deal remains argued over. Its admirers credit it with restoring faith in banks, building infrastructure that lasted generations and creating a floor of security. Its critics argue that some programmes, notably attempts to fix prices and wages through industry codes, slowed recovery, and that full employment came only with wartime spending. Historians also note who was left out. The 1935 Act's definition of employment excluded "agricultural labor" and "domestic service in a private home", which left out many of the poorest workers, including a large share of Black workers in the South; historians still debate how far that was the intent. Both the achievement and the exclusion are part of the record.

Beveridge's five giants

In June 1941 the British government asked the economist and former civil servant William Beveridge to survey the country's patchwork of social insurance schemes. His report, Social Insurance and Allied Services (Cmd 6404), was published at the start of December 1942, in the middle of the war.

Beveridge did not write like a civil servant:

"A revolutionary moment in the world's history is a time for revolutions, not for patching." — Beveridge Report, para. 7

Social insurance, he wrote, was only an attack on "Want". But "Want is one only of five giants on the road of reconstruction and in some ways the easiest to attack. The others are Disease, Ignorance, Squalor and Idleness" (para. 8). Each giant pointed to a field of policy:

Labour won the 1945 election promising to slay the giants; Beveridge himself, a Liberal, lost his seat. The National Insurance Act and the National Health Service Act (both 1946) and the National Assistance Act (1948) followed.

Two things about Beveridge's plan are often forgotten. First, it was meant to be a floor. Its first two "fundamental principles" were a "flat rate of subsistence benefit" and a "flat rate of contribution", with voluntary insurance on top for those who wanted more. Second, it was built on three assumptions it could not deliver itself: children's allowances, comprehensive health services, and "maintenance of employment, that is to say avoidance of mass unemployment" (para. 14). A universal insurance scheme cannot survive mass joblessness, so the welfare state and Keynesian demand management were built as a pair.

Part 4 — The golden age, and the three worlds of welfare capitalism

Thirty glorious years

After 1945 the rich world had the fastest sustained growth in its history. The French call the period les Trente Glorieuses, the thirty glorious years. On the Maddison Project's estimates (in 2011 dollars), output per person grew like this:

Source: Maddison Project Database 2023, computed from GDP per capita. Germany is the unified-territory series.

The framework around this growth was a mixed economy everywhere. The Bretton Woods system of 1944, which Keynes helped design, fixed exchange rates against the dollar and created the IMF. The Marshall Plan (Secretary of State George Marshall's speech of 5 June 1947, Congress's Economic Cooperation Act of March 1948) eventually sent more than $12 billion to rebuild Western Europe. Many governments owned railways, mines and utilities, and unions were strong. And social spending grew steadily:

Source: OECD Social Expenditure Database (SOCX) and OECD historical series, via Our World in Data.

Did the welfare state cause the boom, or did the boom pay for the welfare state? Economists still argue. Much of the golden-age growth came from catching up: Europe and Japan were rebuilding from war and adopting American technology, and catch-up growth is fast while it lasts. The fact that Japan, with very low social spending, grew fastest of all suggests the welfare state was not the engine. The fact that the big spenders grew fast too suggests it was not much of a brake either.

The golden age settled one argument and opened another. A large welfare state turned out to be compatible with rapid growth. Whether it helped is still disputed.

Three worlds

By the 1980s it was clear that rich democracies had built very different welfare states, not just bigger or smaller ones. The Danish sociologist Gøsta Esping-Andersen gave the differences a framework in The Three Worlds of Welfare Capitalism (1990), still one of the most cited books in the field.

His key idea was decommodification: how far a welfare state lets people live decently without having to sell their labour on the market, when they are sick, old, unemployed or caring for children. He found three clusters, using the United States, Germany and Sweden as his typical cases:

The typology is a lens, not a law, and its critics have sharpened it. Feminist scholars argued it ignored unpaid care work done mostly by women at home, which is what Bismarck-style systems quietly relied on. Maurizio Ferrera and others argued that Southern Europe (Italy, Spain, Portugal, Greece) forms a fourth world, with generous pensions and heavy reliance on family. Scholars of East Asia found Japan, South Korea and Taiwan hard to fit at all, since their welfare states were built to serve economic growth rather than to protect people from markets.

Each world has its defenders. Supporters of the liberal model argue that it keeps taxes low, work incentives strong and help targeted at those who need it most. Supporters of the conservative model value its link between contribution and benefit, and its respect for family and for institutions run by employers and unions rather than the state. Supporters of the social-democratic model argue that universal programmes are more popular, harder to cut and less stigmatising, and that high-quality public services make a high-tax country a good place to live and work. Each of the three has prosperous, stable democracies to point to.

Part 5 — Stagflation and the turn of the 1980s

When the recipe stopped working

The Keynesian consensus rested on an apparent trade-off: governments could choose a bit more inflation in exchange for a bit less unemployment. In the 1970s both went up together. Economists called it stagflation.

The Federal Reserve's own history of the "Great Inflation" gives the outline. US inflation, around 1% a year in the early 1960s, ratcheted upwards through the decade. The Arab oil embargo of October 1973 quadrupled crude prices; the Iranian revolution of 1979 tripled them again. On World Bank figures, consumer prices in the United States rose 11.1% in 1974 and 13.5% in 1980. In Britain they rose 24.2% in 1975. Unemployment rose at the same time.

Milton Friedman had predicted something like this. In work cited by the Nobel committee when it awarded him the 1976 prize, he argued that the trade-off between unemployment and inflation was "only a temporary phenomenon": push unemployment below its natural level and people come to expect inflation, so you need ever more of it to get the same effect. The 1970s looked like a demonstration.

The cure, and its price

The turn came from central banks first. Paul Volcker became chairman of the Fed in August 1979, with inflation above 11%. The Fed squeezed the growth of money and let interest rates soar; the effective federal funds rate averaged 19.1% in June 1981. The economy went into a deep recession from July 1981 to November 1982. Unemployment peaked at 10.8% in November 1982, the highest in the official monthly series, which starts in 1948. By the end of the recession, inflation was under 5%. Volcker's defenders say he broke inflation and laid the ground for two decades of stability. His critics point to the millions of jobs it cost, and to the debt crisis that high American rates set off in Latin America.

Politics followed. Margaret Thatcher became British prime minister in 1979 and served until 1990. The UK government's own history of her premiership lists "a radical programme of privatisation and deregulation, reform of the trade unions, tax cuts and the introduction of market mechanisms into health and education", plus the Housing Act 1980, which gave council tenants the right to buy their homes. Ronald Reagan became US president in January 1981. His inaugural address contained the most quoted line of the era, usually quoted without its first four words: "In this present crisis, government is not the solution to our problem; government is the problem."

Here is a fact that surprises people on both sides. The welfare state did not shrink. On the OECD's figures, British public social spending was 16.9% of GDP in 1980 and 17.2% in 1990. In the United States it went from 12.9% to 13.2%. The average across OECD countries rose from 14.4% to 16.5%. What changed was the rest of the mixture: state ownership of industry, the power of unions, the regulation of finance, top tax rates and the goals of economic policy, where controlling inflation replaced guaranteeing full employment.

The 1980s did not dismantle the welfare state. They dismantled the other half of the post-war settlement: state-owned industry, strong unions, controlled finance and the promise of full employment.

The Washington Consensus

In 1989 the economist John Williamson, at what is now the Peterson Institute in Washington, summarised the policy advice that Washington's institutions (the IMF, the World Bank and the US Treasury) were giving to Latin American governments emerging from the debt crisis. He listed ten points:

1. Fiscal discipline
2. Redirecting public spending from subsidies to health, education and infrastructure
3. Tax reform: broad bases, moderate rates
4. Market-determined interest rates
5. A competitive exchange rate
6. Trade liberalisation
7. Openness to foreign direct investment
8. Privatisation
9. Deregulation
10. Secure property rights

Williamson called this the "Washington Consensus". The name stuck, and soon came to stand for much more than he intended, including rapid opening to foreign capital, which was not on his list. In the 1990s versions of the package were applied across Latin America, Africa and the former Soviet bloc.

The results were mixed enough that both sides still cite them. On the Maddison estimates, output per person in Poland, which reformed quickly from 1990, fell for a couple of years and then rose to about four times its 1990 level by 2022. In Russia, it fell by about a third between 1990 and 1996 and took until the early 2000s to regain its 1990 level. Defenders of the reforms argue that the difference lay in how consistently they were carried out and in the institutions underneath them; critics argue that rapid privatisation without those institutions handed national assets to a few insiders. By 2006 the economist Dani Rodrik could write that supporters and critics alike agreed the Consensus had not delivered what it promised, and that the real argument was about what should replace it. His own answer was humility: diagnose the particular constraint holding back each country's growth, rather than apply one list everywhere.

Part 6 — "Neoliberalism": a word worth handling with care

No word in this story is used more, or defined less, than neoliberalism. Anyone who reads about political economy will meet it constantly, so it is worth knowing where it came from and what different people mean by it.

The word is older than most people think. In the 1930s and 1940s it was a label some liberal economists chose for themselves. They wanted to renew classical liberalism after the Depression, which they thought had discredited laissez-faire, and they argued that markets needed a strong legal framework built and defended by the state. In April 1947 a group of them, including Hayek and Friedman, met at Mont Pèlerin in Switzerland and founded the Mont Pelerin Society, whose statement of aims began: "The central values of civilization are in danger."

By the 1990s the meaning had flipped. In a 2009 study, the political scientists Taylor Boas and Jordan Gans-Morse analysed 148 journal articles published between 1990 and 2004. They found three problems: the term was often left undefined, it was used unevenly across ideological divides, and it was applied to an excessively broad variety of phenomena. They traced how it had gone from a positive label coined by German liberals to a negative term associated with radical economic reforms in Pinochet's Chile.

So when you meet the word, ask which of these the writer means:

These are different claims, and they can be true or false separately. Someone can accept that independent central banks beat inflation and still think financial deregulation went too far. The historian Quinn Slobodian, in Globalists (2018), argued that the original neoliberals were never anti-state at all: they wanted strong international rules to protect markets from national democratic pressures. That reading is contested too.

"Neoliberalism" is a real thing to argue about. But it is several real things, and an argument goes better when both sides know which one they mean.

This library uses the word sparingly, and when it does, says which sense it has in mind.

Part 7 — 2008 and after: the state comes back, again

The crash

The Great Recession in the United States ran from December 2007 to June 2009, the longest downturn since the Second World War. The Federal Reserve's history of it records how severe it was: real output fell 4.3%, home prices fell about 30%, the S&P 500 fell 57% from its peak, household net worth fell from about $69 trillion to $55 trillion, and unemployment doubled from 5% to 10%. When the investment bank Lehman Brothers filed for bankruptcy in September 2008, the panic went global.

The response was a vast state intervention, carried out mostly by governments that believed in markets. The Fed cut its policy rate to near zero, supported the insurer AIG the day after Lehman fell, and in November 2008 began buying mortgage bonds and government debt on a large scale. Governments propped up banks and ran large deficits. Friedman's lesson from Part 2 had been learned: nobody let the money supply collapse as it had in the early 1930s.

The austerity argument

From 2010, with debts swollen, many European governments turned to austerity: spending cuts and tax rises to bring deficits down. Supporters argued that high debt would itself damage confidence and growth, and that Greece's crisis showed what happens when lenders lose faith. Opponents argued that cutting spending in a slump would shrink the economy and make the debt burden worse.

In January 2013 the IMF's chief economist, Olivier Blanchard, and his colleague Daniel Leigh published a remarkable admission. Looking back at forecasts made in 2010–11, they found that countries planning bigger cuts had grown less than forecast, and concluded: "A natural interpretation is that fiscal multipliers were substantially higher than implicitly assumed by forecasters." In plain English, austerity had hurt growth more than the experts expected. The debate did not end there. Defenders of consolidation note that the effect weakened in later years and that some countries, such as Ireland, recovered strongly after cutting.

Pandemic, inflation and industrial policy

The pandemic of 2020 pushed public social spending to records: on OECD figures it hit 23.7% of GDP in the United States, 25.6% in Britain and 34.7% in France, before falling back. Inflation then returned for the first time since the 1980s: consumer prices rose 8.0% in the United States, 7.9% in Britain and 8.5% in the euro area in 2022 (World Bank). Central banks raised rates sharply, an echo of Volcker on a smaller scale.

Meanwhile, rich democracies rediscovered industrial policy: the state choosing which industries to support. The US CHIPS and Science Act of 2022, for example, gave the Department of Commerce $50 billion for semiconductors, $39 billion of it for incentives to build factories in the United States. Supporters argue that security, resilient supply chains and the climate transition justify steering investment. Critics argue that governments are poor at picking winners and that subsidies flow to the industries with the best lobbyists. Whether the 2020s mark the end of the post-1979 era or just a correction within it is an open question.

Part 8 — State capitalism: China, Singapore and the sovereign wealth funds

The mixed economies of the West are mostly private economies with a large welfare state. A different mixture, often called state capitalism, keeps markets and private firms but gives the state a much bigger role as owner, investor and director of the economy.

China: the socialist market economy

Capitalism in this series told the outline of China's story, and the next piece examines it as a natural experiment. Here the question is what kind of mixture it is.

From December 1978 China moved step by step from central planning towards markets: agriculture was de-collectivised, state firms were allowed to sell output above their plan quotas at market prices, the country opened to foreign investment, and entrepreneurs were allowed to start businesses. In 1992 the Communist Party adopted the official term "socialist market economy", and in 2001 China joined the World Trade Organization. The World Bank's summary of the result: since 1978, GDP growth "has averaged over 9 percent a year, lifting almost 800 million people out of extreme poverty."

Yet the state never let go of the commanding heights. State-owned firms still dominate sectors such as finance and energy, and the government steers investment through five-year plans, state banks and local governments.

Which part deserves the credit is one of the great live arguments in economics. In Markets over Mao (2014), Nicholas Lardy of the Peterson Institute argued that private firms were the main engine of growth and the main source of new jobs, and found little evidence that state firms had gained ground in the 2000s. In The State Strikes Back (2019), he argued that under Xi Jinping the state was reasserting itself, propping up weaker state firms at the expense of more productive private ones, and that this threatened China's growth. Others argue the reverse: that the state's role in building infrastructure, directing credit and nurturing strategic industries such as electric vehicles and batteries is exactly why China succeeded where many market-reforming countries did not.

China's growth is a Rorschach test. Market economists see what markets did once the state stepped back. Developmental-state economists see what the state did when it stepped in.

Singapore: the state as investor

Singapore is a different kind of mixture. It is one of the most trade-dependent economies on earth: on World Bank figures its exports plus imports came to over three times its GDP in 2024, against about a quarter in the United States. Yet its state also builds much of the housing, runs a compulsory savings system and holds large stakes in major companies.

• The Central Provident Fund (1955) is a compulsory savings scheme: workers and employers pay into individual accounts that fund retirement, housing and health care. It is closer to forced saving than to Beveridge-style insurance.
• The Housing and Development Board (1960) builds public flats, which are sold to households on long leases rather than rented.
• Temasek (1974) holds and manages the government's stakes in companies.
• GIC (1981), "the brainchild of Dr Goh Keng Swee", in its own words, manages most of the government's financial assets.

On the Maddison estimates, Singapore's output per person went from about $3,500 in 1960 to over $80,000 in 2022, among the highest in the world. Admirers say it shows that a lean, disciplined state that invests rather than spends can beat both the welfare state and laissez-faire. Sceptics point out that Singapore is a city-state that grew as a trade and finance hub, that its political system has been dominated by one party since independence, and that its model may not transfer to large, diverse countries.

Sovereign wealth funds

A sovereign wealth fund is a state-owned investment fund, usually built from commodity revenues or trade surpluses. It lets a government save windfalls for future generations and smooth spending across good and bad years, and it makes the state a large owner of private companies.

The best known is Norway's Government Pension Fund Global. After oil was found off Norway in 1969, parliament decided the revenue should be used cautiously; legislation in 1990 created the fund, and the first money went in in 1996. It invests only abroad, so that oil money does not overheat the Norwegian economy. According to its manager, Norges Bank Investment Management, it owns almost 1.5% of all shares in the world's listed companies, spread across about 7,200 firms, and was worth 22,683 billion kroner at the end of June 2026. Budget surpluses flow in and deficits are covered from it.

Will a state investor pursue returns, or political goals? Who controls it? In 2008 a group of funds agreed the Santiago Principles, 24 voluntary principles on governance, transparency and investing on commercial grounds, now promoted by the International Forum of Sovereign Wealth Funds. How much each fund actually discloses varies widely.

The case for and against

The case for state capitalism, in the words of its strongest defenders, such as the political scientist Chalmers Johnson (MITI and the Japanese Miracle, 1982), the economist Alice Amsden (Asia's Next Giant, 1989, on South Korea) and Ha-Joon Chang (Kicking Away the Ladder, 2002): almost every country that caught up with the leaders, including Britain and the United States in their own early days, used tariffs, subsidies and directed credit to build industries that markets alone would not have built in time, and a capable state can discipline firms by tying support to export success.

The case against: the same tools in the hands of a weak or corrupt state produce white elephants and cronyism, and for every East Asian success there are failures elsewhere. State ownership blurs politics and business, so firms are kept alive for political reasons (the "soft budget constraint" of the next piece). And in authoritarian systems, economic power in the state adds to political power in the party.

Part 9 — Social spending compared, and what the numbers can't tell you

Here is how much rich and middle-income democracies spend on public social programmes (pensions, health, family benefits, unemployment, disability, housing support and the like), from the OECD's Social Expenditure Database:

Source: OECD SOCX, as published by Our World in Data (dataset updated March 2025). Recent years for some countries are OECD estimates and may be revised.

Some patterns jump out. The conservative and social-democratic worlds of Part 4 cluster at the top. The liberal world (the US, Canada, Australia) sits near or below average. Countries that grew rich recently, such as South Korea, spend much less but are catching up fast: Korea spent 2.5% of GDP in 1990 and 15.3% in 2024, as its population aged.

Before you use this table to win an argument, four cautions.

1. Age drives much of it. Pensions and health care are the biggest items, and both grow with the share of older people. Japan's spending rose from 3.3% of GDP in 1960 to about 25% now, mostly because it became one of the world's oldest societies, not because it adopted a new philosophy.

2. Public spending isn't all social spending. These figures count public programmes only. Countries such as the United States and Switzerland route much social protection through employers and private insurance, often encouraged by tax breaks, which this measure leaves out. A country can look "lean" on this table and still spend heavily on social protection through other channels.

3. Ireland is an odd case. Much of Ireland's GDP is profit generated there by foreign-owned multinationals that flows straight out to their owners abroad, which is why its statistics office publishes an alternative measure, "modified GNI". Divide by an inflated GDP and almost any ratio looks small.

4. Spending is not the same as results. A high number is not proof of generosity; a low number is not proof of efficiency.

The share of GDP a country spends on social programmes tells you something about its choices. It tells you much less about whether those choices work.

What the table does show is the central fact of this piece: every country in it is a mixed economy. The argument in rich democracies is over the mixture: how much insurance to provide, through what institutions, how much the state should own or steer, and how hard to try to tame the business cycle. Data alone won't settle it, because it also turns on how much weight to put on freedom from want, freedom from taxation, equality and self-reliance.

The takeaway

The mixed economy was built in layers. Bismarck's social insurance (1880s) was a conservative bargain to give workers a stake in the existing order. The Depression produced Keynes's argument that demand can stay too low for years and that the state must fill the gap, and the New Deal and Beveridge's attack on the five giants built a floor beneath citizens. The golden age showed that big welfare states could coexist with fast growth, and Esping-Andersen showed they came in different kinds. Stagflation ended the post-war settlement; the 1980s privatised industry, tamed unions and inflation and deregulated finance, but did not shrink social spending. "Neoliberalism" names several different things and should be used with care. 2008, the pandemic and the return of industrial policy brought the state back in, while China and Singapore show a different mixture again, with the state as owner and investor. Today public social spending ranges from about 10% to over 30% of GDP across the OECD, averaging about 21%. The real argument is over the mixture, not the label.

Try this: pick one benefit you or your family rely on (a pension, a health service, child benefit, unemployment insurance) and trace it back. When was it created, by which government, and in answer to what crisis? Then ask which of Esping-Andersen's three worlds it belongs to: earned by contributions, given to everyone as a right, or targeted at those in need. You will find your country's welfare state is a layered history, not a single design.

Sources & Further Reading

Sources
- Deutsche Rentenversicherung, "Historie" (history of German pension insurance): the 1883, 1884 and 1889 laws, pension at 70 after 30 years, contribution rate, 126,400 pensions in 1891, and Bismarck's motive.
- Franklin D. Roosevelt Presidential Library, "Great Depression Facts": unemployment of 24.9% in 1933, inauguration date, New Deal agencies. US National Archives, "Social Security Act (1935)".
- J. M. Keynes, The General Theory of Employment, Interest and Money (1936), chs. 12 and 24 (Project Gutenberg Australia text). Econlib, "John Maynard Keynes" (biography; Bretton Woods; p. 379 quotation).
- W. Beveridge, Social Insurance and Allied Services, Cmd 6404 (HMSO, 1942), paras. 7–8, 14 and 17 (Internet Archive scan). US Department of State, Office of the Historian, "Marshall Plan, 1948".
- Maddison Project Database 2023 (J. Bolt and J. L. van Zanden), GDP per capita in 2011 dollars; growth rates computed for this piece.
- OECD Social Expenditure Database (SOCX), OECD Social Expenditure 1960–1990 and P. Lindert, Growing Public (2004), as compiled by Our World in Data, "Public social expenditure as a share of GDP" (updated March 2025).
- G. Esping-Andersen, The Three Worlds of Welfare Capitalism (Princeton UP, 1990).
- Federal Reserve History, "The Great Inflation" and "The Great Recession"; FRED unemployment and federal funds rate series; World Bank inflation and trade series; Nobel press release for Milton Friedman (1976); UK Government, "Past Prime Ministers: Margaret Thatcher"; Reagan Library, Inaugural Address, 1981.
- J. Williamson, "What Washington Means by Policy Reform" (1990), Peterson Institute; D. Rodrik, "Goodbye Washington Consensus, Hello Washington Confusion?", Journal of Economic Literature 44(4), 2006.
- Mont Pelerin Society, Statement of Aims (8 April 1947); T. Boas and J. Gans-Morse, "Neoliberalism: From New Liberal Philosophy to Anti-Liberal Slogan", Studies in Comparative International Development 44, 2009.
- O. Blanchard and D. Leigh, "Growth Forecast Errors and Fiscal Multipliers", IMF Working Paper 13/1 (2013); NIST, "CHIPS for America"; Central Statistics Office Ireland, "Modified GNI".
- World Bank, "China Overview"; World Trade Organization, "China and the WTO"; N. Lardy, Markets over Mao (PIIE, 2014) and The State Strikes Back (PIIE, 2019).
- GIC, "Who We Are"; Norges Bank Investment Management, "About the fund" and Half-Year Report 2026; International Forum of Sovereign Wealth Funds, "Santiago Principles".

Go deeper
- 📘 The General Theory of Employment, Interest and Money — J. M. Keynes. Hard going in places, but chapters 12 (on expectations and "animal spirits") and 24 (on social philosophy) are short, vivid and readable on their own. Free text at Project Gutenberg Australia.
- 📘 The Five Giants: A Biography of the Welfare State — Nicholas Timmins. A narrative history of Britain's welfare state from Beveridge to the present.
- 📘 The Three Worlds of Welfare Capitalism — Gøsta Esping-Andersen (1990). The book that gave the field its vocabulary.
- 🌐 "What Washington Means by Policy Reform" — John Williamson. The original ten points, shorter and more cautious than the slogan they became.
- 📘 Globalists — Quinn Slobodian (2018), and A Brief History of Neoliberalism — David Harvey (2005). Two very different histories of the "neoliberal" idea; read them together and compare.
- 📘 The State Strikes Back — Nicholas Lardy (2019). A careful, data-heavy account of the argument over state and private firms in China, from an economist who thinks markets did most of the work.
- 🌐 Our World in Data — Public social spending since 1880. Play with the chart to watch the welfare state grow, country by country.
- 🌐 See also, in this library: Where Public Money Goes and The Great Political Traditions, in Their Own Words (Politics), Capitalism: How Markets, Prices and Private Property Organise an Economy, Socialism: The Many Kinds, From Utopians to Social Democracy, Communism: The Idea, the States, and What Happened and Anarchism: Order Without Rulers (this series), and next, The Great Debates: Can Economies Be Planned, and What Does the Evidence Say?