At a lunchtime meeting on 15 March, Dennis Snower discussed the current crisis in the provision of Welfare State services and proposed three policy measures in response. Snower is Professor of Economics at Birkbeck College, London, and Co-Director of CEPR's Human Resources programme. His remarks were based on his CEPR Occasional Paper No. 13, `The Future of the Welfare State'. Financial support for this meeting was provided by the UK Economic and Social Research Council. The views expressed by Professor Snower were his own, however, and not those of the ESRC or CEPR, which takes no institutional policy positions.

Snower observed that the time is now ripe for a radical reassessment of the Welfare State. Over the 1950s and 1960s there was a widespread perception that the free market could not provide adequate Welfare State services and that government interventions were therefore required. These interventions, however, reduced private incentives to engage in productive activity, restricted personal freedoms, and captured an ever larger share of governments' budgets. In the 1980s the Conservative Revolution rolled back government ambitions to provide for the poor, the unemployed and the sick. As a result, the distribution of income became dramatically more unequal in many countries, but this failed to make the poor more hard-working and self-supporting. Paradoxically, the Welfare State services that suffered least were those captured by the middle class.

The central problem of providing Welfare State services is that their cost has mounted faster than the rate of inflation. This problem, Snower explained, will become progressively worse with the passage of time. The reason is that the output in this sector consists largely and intrinsically of labour input: doctors cannot significantly reduce the time spent with each patient without reducing the level of care; nor can teachers raise class size without reducing the quality of education.

Thus average productivity in most Welfare State services may be expected to grow much more slowly than average productivity in the economy overall. And since wages in the Welfare State services will not fall significantly behind the average wage level, the cost of the Welfare State will rise inexorably relative to the costs of other commodities. But as economic growth proceeds, the demand for health services, education, insurance against poverty and unemployment, etc., will naturally expand. And since the costs of Welfare State services rise relative to the costs of other commodities, society must consequently allocate an ever-increasing share of GNP to Welfare State services.

In response to this problem, Snower made three proposals to reform the Welfare State. The first is to give individuals the option of relinquishing their entitlement to certain Welfare State services retirement pensions, sickness and invalidity benefits, health care, education in return for a tax reduction of 70% of the cost of the services they are expected to require (assessed on the basis of income, age, sex, marital status, etc.). This option would be supplemented by compulsory insurance against sickness, disability and old age. This proposal puts the decision concerning the split between government and market provision of Welfare State services into the hands of the consumers. The reduction in distortionary tax-and-transfer arrangements should gradually generate enough saving to the government through taxation of the new private sector Welfare State provision to permit improved Welfare State provision to the poor.

The second proposal involves providing government loan guarantees for unemployment, education and health. Many of the problems that Welfare State interventions attempt to address have their origins in imperfect capital markets. The hardship from unemployment, for example, is partly due to people's inability to borrow against their incomes in future periods of employment. The reason is that banks usually have trouble collecting the debts from people who change geographical regions, and thus lending on the basis of human capital would encourage default. The government, however, has a comparative advantage in this area, since it is able to trace people through the tax system. Along the same lines, loan guarantees for education and health insurance would help people internalize the relevant costs and benefits of their decisions.
The third proposal is to give the unemployed the option to use their benefits to provide vouchers to the firms that hire them. The vouchers would be worth more, the longer the individual had been unemployed. This Benefit Transfer Programme would be voluntary, so that workers and firms would join only if it made them better off. Firms could qualify for the vouchers only if they could show that each new recruit was an addition to the workforce over the previous six months, or that they were spending the equivalent of the voucher on training for the new employee.

The Programme would involve no extra cost to the government, since the total amount spent on the vouchers would be equal to the amount that would have been spent on unemployment support. It would give the government an instrument for tackling head-on the problem of long-term unemployment. Since the long-term unemployment rate has no significant effect on wage growth, reducing the number of long-term unemployed would not be inflationary. The Programme would also function as an automatic stabilizer: as the economy moves out of recession, the unemployment pool shrinks, thereby reducing the amount spent on wage subsidies.

Finally, Snower argued that these proposals would have to be supplemented by support for those who are either unable to work (e.g. the elderly, the sick or the disabled) or working more productively in the household sector (e.g. some single mothers with infants). Here income tax credits, medical care cover at reduced premiums, and housing benefits are required. Having avoided some of the sources of cost explosion in the Welfare State, the government could then afford to provide a generous safety net for the nation's poor.