Welfare without the state

Europe's welfare states are squeezed between rising demand and the declining readiness of taxpayers to finance it. The result is big deficits, slow growth, and dissatisfaction with the welfare services on offer. Robert Skidelsky recommends a return to the liberal state: detaching welfare from public funding and reducing to 30 per cent the state's share in national income
January 20, 1996

There is widespread agreement that the welfare state needs to be drastically reformed?certainly slimmed down. Designed in the 1940s to protect weakened capitalist economies against the assault of revolutionary socialism, it is now under assault itself. Governments all over Europe are busy chipping away at entitlements and benefits built up since the war. Yet even minor cuts in welfare spending face huge political costs?as Alain Jupp? is discovering in France. The reason is that most people in Europe have come to rely on tax-financed welfare of one kind or other, and governments have failed to discover a political formula which might wean them from this dependence. Only in the US, whose welfare state is underdeveloped by European standards, does an anti-welfare ideology resonate.

Welfare systems differ across Europe, but a common element is Bismarckian compulsory insurance, with the individual, the employer and the state contributing in varying combinations. William Beveridge, father of the modern British welfare state, set out to slay the ?five giants? of Want, Disease, Ignorance, Squalor and Idleness. His famous report (1942) provided a blueprint for a scheme of social insurance against the interruption or ending of earning power: it covered retirement pensions and unemployment benefit; disability and sickness allowances; together with special spending arising at birth, marriage and death. The Coalition and Labour governments rounded off what we think of as the welfare state by setting up ?free? (tax-financed) health and education services. The fifth giant, Idleness, was to be slain by full employment policies.

Beveridge?s commitment to universalism was a decisive break with the Poor Law tradition, whereby state welfare was strictly for the relief of poverty. Beveridge believed the middle classes would only pay for the welfare state if they shared in its benefits. He relied heavily on wartime solidarity and the power of the state to enforce a ?pooling? of risks, across classes and generations. His financial projections took no account of economic growth and rising expectations. Fifty years on, this interlocking scheme of welfare is in disarray?its coherence destroyed by ad hoc additions and subtractions, its assumptions rendered obsolete by social and economic change, its results mediocre or worse; above all, its costs escalating apparently beyond control.

The anti-welfare revolution in Europe is being driven not by hard-faced Social Darwinians but by the ?fiscal crisis of the state??the growing inability of governments to pay for their spending out of taxes. Since the bulk of the spending is on welfare, spending cuts are bound to hit welfare recipients. Since everyone is locked into the welfare state, as recipient or contributor or both, it is hard for reforming governments to avoid the double charge of punishing the poor for poverty and breaking an implied social contract with their citizens.

The champions of the welfare state have a seductive alternative strategy. If European governments were to pursue expansionary monetary policies, unemployment would come down, growth would be faster, and budget deficits would dissolve in higher revenues and lower spending. ?Look after unemployment,? Keynes once wrote, ?and the budget will look after itself.? This would be fine if budget deficits were purely cyclical, as they were in Keynes?s day. But they are now ?structural??products of steady upward pressure on spending, combined with tax resistance. The welfare state is the primary cause of both, and thus contributes directly to the high interest rates which keep growth low and unemployment high. Moreover, it weakens the supply side of the economy by giving people less incentive to work and save. At its extreme, it produces social pathologies summed up in the notion of the ?underclass.?

So we are stuck. The welfare state grows because it is needed and it is needed because it grows. The key to its unblocking lies in the battle of ideas. The choice our communities face is not between more ?welfare? and less growth, or vice versa. It is about how we meet the welfare aspirations of privately wealthy societies: about whether to leave most of the spending on welfare to the private sector or to the state. The key decision (which will haunt the politics of the next two decades) is to uncouple welfare from public finance. Such a momentous shift cannot take place in an intellectual vacuum. It requires the support of political and economic philosophy.

two ideas of the state, the liberal and the collectivist, have struggled for mastery this century. The liberal state is one whose functions are limited (by a ?social contract?) to providing the minimum framework of peace, security and equity by means of which the pursuit of private aims is rendered safe. This allows the provision of ?social safety nets? for the poor, and some equalisation of income and life-chances to create fair starting conditions?and to protect the core values and institutions of a free society: democracy, private property, markets. Liberal thought is impregnated with a profound suspicion of the state, and has been preoccupied with devising a ?constitution of liberty? which gives the state sufficient power to protect people from harming each other, while restricting its power to harm them.

By contrast, collectivism asserts the superiority of public over private purposes, and the right of the state, by virtue of its superior reason, to promote centrally chosen goals. Fascism and communism are this century?s extreme examples of collectivism, but socialism is the genus, of which these two creeds are mutations. Democratic socialists, while not doubting the state?s right to steer society towards its preferred goals, have been constrained by the residues of liberalism which survived in their own thought and by the political systems they inherited. These constraints have not been sufficient, though, to stop a collectivist creep even in societies which remain politically free?as in Sweden, where by 1993 the state was responsible for allocating almost three quarters of national income. The idea that unites collectivists is that the whole is, or can be, greater than the sum of the parts; that government can secure substantively superior outcomes to those which would have resulted from the exercise of private choice. The state is the benevolent, far-seeing actor in the human drama; the people foolish and feckless.

Collectivism, not liberalism, turned out to be the ?God that failed.? The reason is revealed clearly enough in the late Ernest Gellner?s scintillating essay in the November issue of Prospect: liberalism is the only political creed consistent with the lack of a ?supreme? principle which could legitimately adjudicate between (incommensurate) values.? From a sociological point of view, liberalism is what we have left when modernisation has done its work destroying traditional beliefs and allegiances, disrupting settled communities, mixing up races and tribes. This accounts for the fact that socialism has lost the hegemonic position it occupied for much of this century. Its goals presupposed an agreement on ?core values? which could not be achieved, except in wartime, without extreme coercion.

However, we must distinguish between different kinds of liberalism. For Hobbes, the state is needed to stop the ?war of all against all? in the pitiless state of nature. Adam Smith?s view of human nature was altogether sunnier. Men, he wrote, have a ?natural propensity to truck, barter and exchange.? This implies the possibility of co-operative behaviour in the ?state of nature? itself. To be sure, classical economics had its dark side, earning the reputation of the ?dismal science? in the hands of Malthus and Ricardo. Marx was able to transmute their doctrines, via Hegel, into a powerful intellectual assault against market-based capitalism. Despite the great Marxist deviation, though, the important contribution of economics to liberal theory is market optimism: the belief that economic freedom would generate growth in wealth and well-being far in excess of anything previously achieved in mankind?s shackled history.

Political liberals tend to be more pessimistic than economic liberals, which explains present disagreement about the prospects of a post-collectivist world. In liberal political theory the state is the solution to the inherent problems of man?s savagery, always vulnerable to overthrow by the passions it sets out to neutralise. Gellner wonders whether the moral relativism which results from modernisation does not in fact leave liberalism defenceless against creeds which wish to extinguish it. What he ignores is that the encouragement liberalism gives to the pursuit of private goods itself tends to tame collective passions. The weakness of the ?agonistic liberals? of our day is that they do not feel in their bones the powerful economic promise of the market system. The world they look at is dominated by political, ethnic and religious strife, legacies of deep histories. They think of the market system as a potent source of the strains which fuel non-economic passions. They often believe that the state should regulate market forces. For economic liberals, by contrast, the market system is inherently co-operative. It does not of itself guarantee peace and prosperity. But it creates a powerful support system for a liberal constitution.

Historical ?states of nature??such as the long epoch which followed the collapse of the Roman Empire, and the rise of the modern nation state?were by no means the lawless jungles imagined by Hobbes. There is thus no compelling reason to believe that the shrinkage of the state as we now know it will lead to a ?war of all against all.? Furthermore, Mancur Olson and others have shown that the public goods we most value can be, and have been, provided by private individuals or voluntary co-operation. Public choice theory has demonstrated what we all knew: that there is no necessary connection between the goods governments provide and the goods people want. None of this means that the state will or should disappear. We are stuck with states. The task is to deflate their pretensions, while preserving their power to do what states need to do.

it is against this revived liberal perspective that the claim of the British state to dispose of 43 per cent of our national income should be scrutinised. This percentage, it should be noted, is almost 10 percentage points higher than it was in the 1960s (the ?golden age? of welfare capitalism), 20 percentage points higher than in the interwar years, and over 30 percentage points higher than at the turn of the century. The ?rolling back of the state? begun by Margaret Thatcher in the 1980s has dislodged the British government as owner of industry, but has merely held steady the rise in state spending. In most of western Europe it has gone on rising. General government spending had risen to an average of 55 per cent of GDP in the main European economies by 1993, compared with 46 per cent in 1979. Tax receipts lagged behind, at 48 per cent of GDP in 1993 and 42 per cent in 1979.

Has the cost of goods and services which, by common agreement, the state should provide, risen by anything like the amount needed to explain this inexorable rise in the proportion of national income spent by the state? The answer is that it has not. In the UK, state spending on public administration, debt interest, law and order, external services and defence?the traditional functions of the liberal state?came to 7 per cent of GDP in 1903 and 13.1 per cent in 1979. The largest single increases were in debt interest and defence?understandable in an era which saw two world wars and a ?cold war.? But by far the largest part of the increase in the state share of national spending since 1903 is attributable to the growth of welfare. Public spending on welfare services (social security, plus health care and education) rose from 3 per cent of national income in 1903 to 24 per cent in 1979. Since the welfare state takes the lion?s share of public spending?nearly 70 per cent of a total of about ?300 billion?attention has naturally come to concentrate on the justifications for this slice of public spending.

Let us start with affordability. From a public finance point of view, a government can afford any level of public spending it can pay for in taxes. Notions of ?prudent? public finance, of ?balancing the budget? rarely go beyond this definition of affordability. However, even on this criterion European governments cannot afford their present level of spending. Only a handful of governments managed to balance their budgets at the height of the 1989 boom. Since the 1960s few western governments have been covering their current spending by taxes. They have ?balanced their budgets? by borrowing and inflation.

Why should this matter? Because continuous budget deficits to pay for state consumption raises the cost of capital, and thus slows down the rate of investment and economic growth. The rate of interest is also pushed up by fear of government default through inflation, as happened in the 1970s. Holders of government debt will demand the premium of higher interest rates against the risk of being paid back in devalued currency. Current interest rates on government debt?the marker for all other rates of interest?are already 2 per cent higher on average in the UK than in Germany, which has a lower deficit and a stronger reputation for sound finance.

If governments try to pay their way by borrowing from the banking system (printing money) they impose a tax on savers and lenders?the ?inflation tax.? If they borrow from the public, they impose a tax on investors and borrowers?the ?interest rate tax.? The first tax is collected through rising prices, the second through rising unemployment. In the 1970s governments relied on the first, in the 1980s they relied on the second. Both are bad for economies. The point is that, in today?s world, it is not unemployment which causes budget deficits; it is budget deficits which cause unemployment, by making much potential commercial activity unprofitable. Inflation and unemployment can be seen as tax-raising mechanisms by governments which cannot balance their accounts.

To this there is an important caveat. Keynes demolished the view that there is a fixed lump of savings, so that any increase in government borrowing is bound to ?crowd out? an equivalent amount of private investment. He said this would be true only at a time of full employment. However, he did recognise that an increase in government borrowing could raise interest rates even in a recession if government finances were regarded as unsound. It was crucial to Keynes?s strategy of keeping interest rates low that governments should run budget surpluses in good times to redeem the debt they had incurred in recessions. Today the markets will not lend to deficit-prone governments at low interest rates?which helps explain why recovery from the recession of 1990-91 is so weak. So the constant upward pressure of public spending meets the unwillingness of savers to finance it except at rates of interest which slow down recovery and thus reinforce the pressure for more spending on welfare.

How do we escape from this bind? One way out is higher taxes. Higher taxes to pay for the welfare state is advocated by many socialist economists, though not by leaders of the Labour party. There are two types of objection to this policy. The first is an economic/moral one. High taxes diminish individual liberty and business incentives, while creating social pathologies spawned by the welfare state. The second argument is an economic/technical one. For example, tax-absorbing ?pay as you go? state pension schemes?in which current contributions are used to pay pensions?slow down the formation of capital and become quite unsustainable in ageing societies. Similarly, in today?s global economy the corporate and individual tax base has become increasingly mobile; and thus tax resistant. Most western societies are already at?or beyond?the limit of taxable capacity. The recourse of the British government to a national lottery to raise revenue for heritage, sport and the arts spending is one response to tightening revenue restraints.

There is no such thing as an ?optimal? tax rate for all societies at all times. In 1945 Colin Clark argued that 25 per cent of the national income was a safe upper limit, beyond which various kinds of tax resistance started to develop. Since then tax tolerance has undoubtedly increased, probably as a result of the depression of the 1930s and the second world war. In general, the feasible tax ratio depends on the efficiency of the tax collecting system, the size and structure of the revenue base, the degree of respect for the law, the efficiency with which state services are provided, and many other things. Tax resistance may develop at quite low levels of taxation if taxes are inefficiently or inequitably levied, or if the goods and services they buy are not what people want. That is why one cannot argue a priori that because the British state?s tax take is lower than, say, the German, the British state has no taxing problem.

But the assertion that British governments have a taxing problem does seem to fly in the face of opinion poll evidence. A 1995 Guardian ICM poll showed 60 per cent of voters wanting the basic rate of income tax increased to pay for higher spending on schools and hospitals. Yet most politicians regard espousal of higher taxes as electoral death. Their caution is supported by electoral experience and other opinion poll evidence. A survey in 1986 showed that 61 per cent of a sample of 2,000 regarded their own household?s income tax as too high; only 3 people in the whole sample thought it too low. So British people seem to want more spending on welfare without being willing to vote the taxes to pay for it. Is this irrational? Not necessarily.

The demand for more public spending should simultaneously be a demand for higher taxes. But when the direct link between a good and its cost is severed, so that an individual can vary his consumption of the good (or at least try to) without having to vary his contribution to its cost, an inconsistency develops between willing the end and willing the means.

Alternatively, the apparent inconsistency in the opinion polls may arise because people want to spend more on welfare but fear that in the hands of the state their money will not be spent for the purposes they want or in ways which they want. If voters have learnt by experience that their spending priorities are not met by the state, it would be irrational for them to give governments even more control than now over their household budgets. The higher the tax burden, the greater the risk that the spending of national income is likely to diverge from the wishes of those who pay the taxes.

?Rational? tax resistance underlies the ?fiscal crisis? of contemporary welfare states. Over many years governments have squeezed spending on services and benefits of greatest concern to the majority?particularly pensions and education?in order to make room for the expansion of those of greatest interest to minorities. The fastest growing components of public spending in the UK since 1979 have been legal aid, invalidity benefit, housing benefit, mobility allowances, attendance allowances and one-parent benefits. Some of this expansion has been driven by deteriorating economic conditions, but most of it has been caused by the ?entitlements explosion? which started in the 1960s.

The essence of these post-Beveridge entitlements is that a commitment is made to provide money to people who satisfy certain conditions irrespective of contribution. Once an entitlement to benefit is established, the population claiming the benefit expands, regardless of economic conditions. Such commitments are virtually open-ended, and over time they tend to ?crowd out? the earlier commitments to provide universal services or benefits in return for taxes or contributions. The majority come to feel they are not getting value for money for their taxes or contributions. They want more spent on the services and benefits of most concern to them, but do not want to pay for the entitlements of non-contributors, many of whom they consider undeserving.

Tax resistance stems not just from the divergence between public and private spending priorities, but from dissatisfaction with the type of service being provided. Wealthier societies want not just more of the same, but more variety. But bureaucracies are best at providing standardised goods. In state education the trend over many years has been towards greater standardisation and centralisation. Variety in type of school, sources of school finance, teaching methods, syllabuses and qualifications have been drastically reduced over time. Conservative governments, while preaching ?variety and choice? have carried a long process to a logical conclusion by nationalising the curriculum and setting ?national attainment targets.? Labour would be even more collectivist than the Conservatives.

at this point, the deficiency in the original plan becomes apparent. Beveridge argued that the basis of the welfare state should be social insurance??benefits in return for contributions rather than free allowances from the state.? However, he rejected the private insurance principle of funding individual risks in favour of pooling risks across classes and generations. In rejecting the principle of contribution tailored to need and preference, and in relying instead on the power of the state to enforce social solidarity through the tax system, Beveridge lit a slow-burning fuse under his own edifice, as the state?s power to compel inter-generational and cross-class transfers shrank.

The high tax parties might try to overcome tax resistance by the device of ?earmarking? or ?hypothecation?: for example, a ?health tax? might be raised to be spent on the NHS; an ?education tax? for schools, thus establishing a closer connection between contribution and benefit. Such a device might ease the tax constraint, but it would be hard to convince voters that the proceeds of the tax would be spent in improving the benefits?spent, that is, on better facilities and services rather than on higher pay and less work for teachers, doctors and so on. Both experience and theory suggest that the lion?s share of tax revenues will go to the producers and not consumers of the services they are intended to pay for. And earmarked taxes will not solve the problem of benefits untailored to individual requirements.

An alternative to raising ?earmarked? taxes is to re-balance public spending within slimmed-down totals. Kenneth Clarke?s November budget statement started on this by increasing allocations for health and education within a reduced public spending target. This was an attempt to re-order spending priorities in line with majority concerns. However, this cannot go very far without squeezing the means-tested benefits, thus hitting the most vulnerable. The opposite method of re-balancing would be to reduce middle class entitlements and target public spending on those in greatest need. For example, universal child benefit might be abolished, and the better-off made to pay more of their own education and medical bills. In this model the welfare state reverts to a ?social safety net.? But, despite the widespread support for more targeting, it runs directly counter to the other widely shared concern: to abolish or reduce means-testing, with its institutionalisation of dependency. Abolishing universalism also destroys the implicit Beveridgean social contract which locked the better-off into the welfare state.

Faced with these unenviable choices, politicians are likely to settle for chipping away at the spending without attempting any radical restructuring of the system. By strenuous (possibly heroic) efforts, helped by a favourable rate of economic growth, it may be possible for the British state to balance its taxes and spending at about 40 per cent of GDP by the end of the century. The disadvantage of this approach is that it will yield few, if any, tax reductions, will be highly vulnerable to ?reform fatigue? and any interruption to growth, and will still leave the state in control of too much of the national income.

my own position should by now be clear. It is that western states tax and spend too much for the health of a free society and free economy. Many things which people could do for themselves, and might well want to do for themselves, are now being done by the state. There is no reason, for example, why education spending in Britain should be limited to 5.6 per cent of national income. This is purely a public finance limit: it has little relation to what parents might wish to spend if they had greater control over their household budgets. In fact a majority of parents want more spent. Fifty per cent say they would send their children to expensive private schools ?if they could afford to.? At the same time, the huge share of national income pre-empted by welfare spending, and the financing problem to which this gives rise, prevents the state from doing some of the things which modern states should do. For example, governments cannot use budgetary and interest rate policy to smooth out the business cycle if their deficits are permanent and their debts growing. Our transport infrastructure has been shamefully run down?squeezed out by social spending. Yet it was Adam Smith who saw the maintenance of infrastructure as one of the defining duties of the state.

I believe that the share of national income raised and spent by the state should be reduced to about 30 per cent. This figure is not picked out of thin air. It is close to the government share in the dynamic Asian economies. It is argued that the social structures of these societies are very different from our own, but they are not so different from what ours used to be before welfare was nationalised. We can learn from them in other ways: Singapore, for example, operates a funding system for its social security. Its Central Provident Fund is a state-operated compulsory savings scheme, in which contributions from employers and employees are held in individual accounts, a portion of which employees can invest in stocks or housing. Investors get out of the programme exactly what they put in, plus any returns, to fund retirement or hospitalisation costs. By investing in bonds the CPF financed most of Singapore?s infrastructure investment of the 1970s, but today an increasing share of its investments are made in private business. The Labour MP Frank Field has argued for a similar plan in the UK. Schemes of this kind offer a half-way house between compulsion and voluntarism. They restore the link between individual contribution and benefit, while ensuring a high savings ratio.

The 30 per cent share is also roughly that spent by western governments in the ?social-democratic? 1950s and early 1960s?when growth rates were historically high and before the familiar symptoms of tax resistance appeared. A return to such levels would not be a return to the 19th century state, but to the successful phase of the post-war welfare state.

The principles underlying a ?British state reduction programme? might look like this. It would be directed at weak claims rather than weak claimants. There is no reason why an increasing share of spending on health care, education and retirement pensions should not come from private budgets. This is consistent with universalism, provided it is understood that state spending will be a decreasing share of total spending. Public spending on health and education could be linked to the rate of inflation with ?top-ups? coming from the rise in real private earnings, in the form of co-payment for school fees, more private insurance for health care, grants and bursaries from banks, and charities. Such a public spending strategy would make possible progressive reductions in taxation, down to 30 per cent. Government spending could be redirected towards the poor without any increase in means-testing if it was ?weighted? by locality rather than by test of individual means. The result would be a more heterogeneous welfare system, with payments from private households, the voluntary sector and the state. The promise of such a programme is that it would rejuvenate our sclerotic society, and have positive effects on economic growth?easing the pains of the transition. It bristles with technical and political difficulties but it represents the unfinished half of the Thatcher revolution.