Two decades of government effort to 'roll back the frontiers of the state' in welfare, constrained by both economic and political factors, have resulted in a complex mix of public and private sector activity. Tania Burchardt, John Hills and Carol Propper explored the development of private welfare and the role it currently plays in education, health, housing, income maintenance and personal services, to tease out the opportunities for, and limitations on, further attempts to change the boundaries between public and private welfare. The study found:
-Welfare has never been the exclusive preserve of the state. Private involvement has taken many different forms in the provision, finance and control of services. Already in 1979/80, nearly half (48 per cent) of all welfare spending was on services with some private involvement.
-Shifts in the welfare mix have been slow. By 1995/96, just over half (51 per cent) of all welfare expenditure was on services with some private involvement. Within this total, privately provided services have grown faster than privately financed services.
-Differences in attitudes towards public finance for welfare between those who make use of private services at a particular moment in time, and those who use public services, are less than might be expected, perhaps because of the complexity of public/private mixes in provision and service use.
-The current combination of finance for, and use of, welfare services is highly redistributive. Moves towards private finance for even relatively small parts of the welfare package could have dramatically regressive effects.
-Given these and other constraints, the authors conclude that, although the search for an efficient mix of public and private welfare provision is likely to continue, opportunities to shift from public to private finance are limited, and are likely to prove costly in major areas of welfare.
Key questions and concepts
In its April 1998 Green Paper on welfare reform, the new Government argues that its policy reforms will follow a 'third way':
'The welfare state now faces a choice of futures. A privatised future, with the welfare state becoming a residual safety net for the poorest and most marginalised; the status quo, but with more generous benefits; or the Government's third way - promoting opportunity instead of dependence, with a welfare state providing for the mass of people, but in new ways to fit the modem world.'
The most radical forms of privatisation have been ruled out, as has the possibility of doing nothing, but where exactly the boundary between public and private sectors in welfare should lie remains to be determined. This research sets out the background to the debate and provides empirical evidence on the scope for further private involvement in welfare activity. The central questions addressed are:
-How can the complexity of welfare activity best be understood? How has the mix between public and private sectors developed? -Who uses private welfare services and why? Are they are an exclusive group or do people 'mix and match' different services at different times? -In what circumstances do the efficiency arguments in favour of private provision hold? What effect would a shift towards greater private finance of welfare services be likely to have on different parts of the population? -Where are there opportunities for moving the boundary between public and private welfare and what are the constraints?
Changes in boundaries between public and private welfare
A simple distinction between 'public' and 'private' is unhelpful in thinking through policy options. Instead welfare activity was analysed on three dimensions (see Figure 1)
-Provision: is the provider a public or private sector body? -Finance: does the public sector pay for the service either directly through subsidy or indirectly through benefits or tax relief? -Decision: can individuals choose for themselves the provider used or the amount of service received?
This three-way classification, with public and private options within each, generates eight possible combinations, such as public provision with private finance and decision (light purple sector; for example, pay beds in NHS hospitals), or private provision with public finance and private decision (light green sector; for example, tax reliefs on additional voluntary contributions to a pension scheme).
Welfare activity was already very mixed in its composition in 1979/80, with the 'pure public' sector - publicly provided, financed and determined - only making up 52 per cent of the total.
Shifts in the welfare mix have been slow (see Figure 2). The private sector has become more important, but the pure public sector still represented nearly half of all welfare expenditure in 1995/96. Expenditure on welfare services that were privately provided rose
from 41 to 49 per cent of total welfare expenditure between 1979/80 and 1995/96, while expenditure on privately financed services rose from 27 to 31 per cent of the total.
Patterns vary between areas of welfare: housing started with a large privately provided and financed sector and this grew over the period, while in income maintenance (including social security) the 'pure public' sector increased its share. Personal services appears to have made most use of contracting-out arrangements. In both health and education the 'pure public' sector remains dominant, but the 'pure private' sector grew appreciably.
These changes have taken place within a growing overall total: public spending on welfare grew in relation to GDP between 1979 and 1995. Overall welfare activity - including privately financed services and expenditure not usually counted in public spending figures, such as tax reliefs - grew even faster. The pure private sector - consumers choosing to pay out of their own pockets for privately provided services - was£83 billion by 1995, nearly 12 per cent of national income. (Some of this went on services also supported by public finance, so an even smaller proportion of spending is entirely free of any connection with the state.)
Who uses private welfare?
The use of key forms of private welfare has been growing, particularly private medical insurance, dental and eye care, private pensions, and, of course, owner-occupation. It has both broadened - more people use it - and deepened - those who use it
-Growth has occurred in all income groups, although use of private welfare is at a much higher level for those with high incomes. -Average spending by those who pay for private schooling, medical insurance and pension contributions has doubled (or nearly so) since the 1970s, over and above inflation.
Patterns in service use are generated by the preferences consumers have and the constraints they face (for example, if eligibility for 'free' public services is restricted). As a result of this combination, use of a private service in one year is associated with using it again the next - for example, just over half those who use private dentistry this year go on to use private dentistry in the next. However, nearly a third of private users in the first year switch to using only the NHS in the next. Use of private dentistry is therefore strongly associated with repeated use, but it is by no means a one-way street.
In education, there is evidence of an intergenerational effect: two-thirds of those households where both parents went to private school have a child who went to private school, more than 10 times the proportion of households where neither parent went to private school. Again, however, for many children private schooling is only one part of their total education, rather than representing the exclusive input. A very large proportion of those using private secondary schools - 85 per cent in 1997 - go on to (state-funded) universities, and according to a MORI survey, just under half of pupils in their first year at private secondary schools had attended a state primary school.
Individuals who use a private service in one area of welfare are also more likely to make use of another. However, many users of one private service (particularly housing) do not use any others and only a small minority use the full range of private welfare services in a year: 11 per cent use all three of private housing, health and pensions, and just 5 per cent use private housing, health and education. Even fewer use all four private services.
To summarise, the evidence suggests that although users of private welfare tend to have certain characteristics - for example, higher incomes, higher educational qualifications, and support for the Conservative Party - they do not form a separate and distinct part of the population. There is not a sizeable group who exclusively use private services: most users of private welfare also use public services as well or at other times, and often their use depends in part on public finance.
Attitudes towards state welfare
The fact that a majority of users of private welfare services also use publicly provided or publicly financed services may help to explain the relatively small differences in attitudes towards public services expressed by 'private' and 'public' users. At a general level, increased use of private welfare has not been associated with a fall in support for tax-financed welfare services; indeed, support for additional state spending on health and education has grown over the same period.
At an individual level, those with children who have used private education are not significantly less in favour of higher state education spending than others when other factors, such as income and educational qualifications, are taken into account. There is, however, some indication that users of private health care are slightly less supportive of universalist principles, even when their other characteristics (such as income and political affiliation) are controlled for. This could mean greater use of private health care would undermine support for the NHS, but analysis of national survey data reveals no evidence that using private health care leads in itself to a change in attitudes. If anything changes attitudes towards the NHS, it seems to be use of health care at all, not whether that use is public or private.
Constraints and opportunities for private welfare
Reasons for using private welfare services are generally much more pragmatic than ideological. As in any other market, given the particular structure of and constraints on what is available from the state and the private sector, people make decisions which best meet their own preferences given what they can afford. Some common situations are illustrated in Figure 3.
(a) No state-financed provision. The higher people's income the more they tend to buy, but there is a range (shaded) for any particular income level between those who value other items more or less.
(b) Private additions to state provision. The state provides a universal flat rate service of the same value to all, but people can add to it privately. The effect of state intervention is to set a minimum level of provision below which people do not fall.
(c) Private services as a niche market alternative to state-financed provision. Those using the private sector have to do so instead of using the flat rate state provision, while still paying their taxes in full. This does not affect those who are brought up to the minimum. But those who would otherwise have chosen more than the flat rate amount, have to purchase the whole amount from the private sector to increase the amount of the service received.
These simplified cases bring out the important difference between a private service as an addition or an alternative to what is available from the state. In the former case one might expect quite a lot of people to be combining a fairly small amount of private spending with use of the basic state service - the private sector might be broad, but not very deep. In the latter case, one might expect a relatively small number of people to be spending quite a lot privately - the private sector will be deep rather than wide.
The case where the private sector is an alternative to a public service can lead to even stronger pressures from some private users to increase public spending, but from others to reduce it. Those who remain with a public service which is somewhat below what they would like may well favour a significant increase in public spending both to achieve the level of service they want, and to avoid the danger of having to make an expensive decision to opt out. On the other hand, those who have opted for private services which are well clear of the public level will see little return for themselves from greater public provision and may oppose it.
Removing publicly financed provision altogether resolves this conflict but could have dramatically regressive effects, even if restricted to relatively small parts of the welfare package. Moving from tax-finance to a flat-rate charge for schooling, for example, could cost households with school-age children in the bottom tenth of the income distribution as much as£3,500 per year. The current combination of patterns of finance for, and use of, welfare services is highly redistributive (Figure 4).
Direction of future policy
The Government's Green Paper on welfare reform talked about the need to increase the use of partnerships between public and private sectors. The analysis in this report suggests that we are unlikely to see just one approach - rather solutions need to be tailored to each area of welfare, taking into account existing patterns of service use and the different effects of varying the combination of public and private involvement. Changing the finance of services is likely to have a significant distributional impact, whereas switching from public to private provision is largely a matter of comparative efficiency. Equity and efficiency objectives affect the choice of mechanism for determining who gets how much of a service.
In education, health, and much of income maintenance, the role of public finance and public decision will remain strong, although there may continue to be increased use of non-public providers for ancillary health services and the pure private sector may continue to grow as a supplement to public provision. By contrast, for housing, the role of public provision and perhaps also public finance may diminish. Private provision of pensions looks likely to continue to grow but subject to stronger public control in terms of regulation and the pensions people are expected to accumulate for themselves.
In one sense, these kinds of development would simply continue trends we have already experienced since the early 1980s. Those trends themselves represent a far more pragmatic reaction to the relative strengths of public and private sectors than ideologues on either side of politics might have argued at the time. The process of establishing in each case whether, and in what combination, the private and public sectors can provide services most efficiently, and finding ways (where information problems and equity objectives allow) to promote individuals' control over the services they receive, will and should continue. But shifting from public to private finance is regressive and, in many cases, is likely to result in only limited savings for government, much outweighed by additional costs
About the study
The research was based in part on analysis of three nationally representative household surveys, supplied by the Data Archive. They were the British Social Attitudes Survey (for 1993 and 1994), the British Household Panel Survey (for 1991-1995; the survey follows a sample of 10,000 individuals and is run by the ESRC Centre on MicroSocial Change at the University of Essex), and the Family Expenditure Survey (for 1979, 1985 and 1995/6). Aggregate statistics on welfare activity and expenditure were drawn from a wide range of government and independent sources.
How to get further information
A more detailed report, Private Welfare and Public Policy by Tania Burchardt, John Hills and Carol Propper, is published for the Foundation by YPS. (price£13.95 plus£1.50 post and packing, ISBN 1 902633 00 8).
A background discussion paper, Boundaries between public and private welfare: a typology and map of services by Tania Burchardt, is available from the Centre for Analysis of Social Exclusion, London School of Economics, Houghton Street, London WC2A 2AE, Tel: 0171 955 6679, where the authors can be contacted.