+44 (0)20 7799 6677

Stakeholder Health Insurance: Empowering the Poorest Patients

by David G. Green
Published in the British Medical Journal, 31 March 2001


The NHS does not achieve its own objectives. It suffers from three long-standing flaws: it is underfunded because its method of funding is unrelated to personal demand and need; there is a lack of competition; and there is a lack of respect for individual choice. Above all, it fails the poorest members of society. A system of stakeholder health insurance would make use of the market to serve everyone.


The NHS aims to be universal, comprehensive, equal and of a high standard but fails to achieve its own objectives. Are there any overseas systems that guarantee all their people – especially the poorest members of society – a higher standard of care than the NHS? An honest observer looking at countries such as Germany, France or the Netherlands would have to answer this question with a resounding ‘yes’. Can we learn from them?

One advantage of European social insurance is that the standard and range of cover is linked to judgements about affordability made by people who are self-sufficient through work. In Germany and France, individuals can identify on their pay slips how much they are paying and form a view about whether the cost is justified. The state guarantee reflects personal preferences to a greater extent than under a system financed from taxes. Put another way, the French and Germans have found a way of making the market serve everyone.

However, the National Plan for the NHS dismissed overseas alternatives because they failed the test of ‘equity’, but in which European countries would you find the widest gap between the standard of care enjoyed by the poorest people and the standard enjoyed by the rich? The people with most to lose are those who depend exclusively on the government for services. The wealthy can always take care of themselves. In countries such as Germany, France and the Netherlands the high standard of care guaranteed by the government narrows the gap between rich and poor; but in the UK the lower standard of care has its most serious impact on the poorest people.

What counts as ‘comprehensive’ at any moment is in the process of being discovered and rediscovered. The advantage of a system based on insurance is that it allows gradual evolution towards a reasonable standard which reflects consumers’ judgements about the type and cost of cover they want. The allocation of funds by the UK Treasury is crude by comparison. It is what the government can afford or chooses to spend and bears little or no relationship to medical demand. In any event it is a global amount with no room for individuals to pay for more or less.

Opponents of insurance typically highlight two main problems: the exclusion of people with pre-existing conditions and the related tendency of some insurers to ‘select’ customers in order to avoid those most likely to make large claims. Over the years many different solutions to these problems have been attempted, but perhaps the most promising have been schemes based on group insurance with a ‘sponsor’ acting as a consumer champion, also called ‘managed competition’. Sponsoring agencies, which could be private organisations, including large employers, or statutory bodies insulated from the political process, facilitate consumer choice by offering comparative information about quality and price and by filtering out bad insurers. Such schemes have been championed for more than 20 years by Professor Alain Enthoven. Originally called consumer choice health plans, the latest name is the health insurance purchasing co-operative. Such systems capture the strengths of European social insurance without some of the weaknesses.

There are four main elements. First, each year consumers choose a comprehensive care package for one year. Second, they do so through agencies whose task is to facilitate choice by providing comparative information about quality and price and by weeding out unsatisfactory insurers. Third, the consumer’s choice should be cost-conscious, that is, part or all of the cost of the premium should be met by all individuals except the absolutely poor. And fourth, providers should compete in structures which integrate provision and insurance, either by establishing a single system, such as a health maintenance organisation or by creating schemes based on contracts between insurers and independent providers.(1) Schemes based on ‘managed competition’ have been found to work. There are examples of such systems in operation in California and Minnesota, but I will mention only one: the Federal Employees Health Benefits Program.

The latter began operation in 1960 and now offers nearly 400 insurance plans to some four million policy-holders covering about nine million people. Every year in November/December there is a month-long ‘open season’ when people choose their insurer for the next year. Insurers must accept all applicants regardless of pre-existing conditions and the federal government typically pays 75 per cent of the actual premium of each person’s chosen plan. Employees pay the difference.(2)

How could such a system be implemented in the UK?(3) Existing health authorities could establish stakeholder health insurers in their areas. We would all continue to pay taxes as at present and health care would continue to be provided through primary care groups, without further charge. People who prefer to be covered by insurance would opt to receive care through their local stakeholder. In return for assuming responsibility for part of the cost, they would receive a tax credit representing part of the tax they had paid towards the NHS.

The stakeholders should be independent of government, and preferably mutual organisations run by boards representing members. Each year, stakeholders would invite private insurers to submit tenders for a comprehensive package of cover for anyone within the stakeholder boundary. All insurers should be required to price a standard contract: to facilitate value-for-money comparisons; to guarantee no hidden gaps in coverage; and to prevent risk selection from reducing incentives to produce value for money. This standard package should be defined by each stakeholder to reflect members’ preferences and to facilitate comparisons.

Hospitals, NHS or private, would charge insurers for their services. Private hospitals, whether for-profit or not, would compete on equal terms. All hospitals would be free to enter into contracts or arrangements with insurers as they believe best. Similarly, GPs functioning through primary care groups would charge insurers or offer pre-paid services.

Initially the scheme would be based on existing health authorities, but as under Enthoven’s scheme, it should be possible to establish mutual purchasing agencies other than area-based stakeholders. A common objection to small-scale purchase of health insurance is that the administrative costs tend to be very high. However, the RAND Health Insurance Experiment found that groups of 10,000 or more have administrative costs of 5.5 per cent, whereas for smaller groups it can be 40 per cent. Thus, groups of 10,000 are large enough to secure the relevant economies of scale.

How might a system of tax credits work? For each person opting to receive insurance cover through the stakeholder rather than the NHS, a sum of money would be paid by the government to their stakeholder. There would need to be an interim arrangement until enough experience had been gained of the evolving insurance market. Two years would probably be sufficient, and during these two years the Treasury should apportion an age-weighted amount per person based on the previous year’s NHS expenditure (approximately £800 per head). In subsequent years, the Treasury allocation should be based on the market price for the standard package defined by each stakeholder. The Exchequer subsidy should be a percentage of this market price.

There would be an individual policy and a family policy. Based on expenditure in France and Germany, the cost might be £1,300 for an individual and £2,600 for a family. Let’s assume that the cost of a standard insurance package for a husband, wife and two children is £2,600 per year. For people on benefit the government would pay the full amount to the stakeholder. The tax credit for people with earnings would taper away at 25p for every pound of tax liability above each family’s tax threshold. The taper could stop when the amount of credit was equal to 50 per cent of the cost of the standard plan. No one would receive more than a 50 per cent subsidy.

For families with no tax liability the stakeholder would receive £2,600 from the government and there would be no out-of-pocket payment. For a family with a tax liability of £1,000 the stakeholder would receive £2,350 from the government and £250 from the family. For a family with a tax liability of £2,600 the stakeholder would receive £1,950 from the government and £650 from the family. A family with a tax liability of £5,200 would pay £1,300 to the stakeholder which would receive £1,300 from the government. At this point the tax credit would be 50 per cent of the standard plan, the maximum subsidy.


The end result would be universal access to a guaranteed standard, rather than universal access to a politically-determined standard which bears little relationship to either national wealth, medical need or personal demand. There would be competition to create room for experimentation and the discovery of new and better ways of meeting human needs. Above all, the poorest people in the society would have been empowered. They would be free to receive care from the NHS as at present. If they preferred to switch to an alternative insurer, they would enjoy the same power to do so as anyone else.


1. Enthoven, A.C., Theory and Practice of Managed Competition in Health Care Finance, Amsterdam: North Holland, 1988; Enthoven, A.C. and Singer, S.J., ‘A single-payer system in Jackson Hole clothing’, Health Affairs, Spring 1994, pp. 81-95.

2. Office of Personnel Management, Federal Employees Health Benefits Program, A Handbook for Enrollees and Employing Offices, 1999.

3. A fuller version of this proposal can be found in Green, D. G., Stakeholder Health Insurance, London: CIVITAS, 2000.


Keep up-to-date with all of our latest publications

Sign Up