What can be learned from the evolution of these different systems?

Each country’s method of revenue raising is the product of policy evolution unique to that country, but there are some common themes that can illuminate some of the benefits and challenges associated with different funding models.

What are the perceived strengths and weaknesses associated with different funding models?

In the UK, debates about the likely benefits and drawbacks associated with moving to a SHI system date back at least two decades. Proponents of changing to social insurance often focus on the additional revenue that could be raised, alongside the supposed advantages of the institutional arrangements that typically evolve in these systems. These include non-governmental bodies raising and spending funds for health care, and particularly their separation from central government, and the assumption that these systems are insulated from the politics that accompany the disbursement of tax-derived funds. Researchers at both the World Bank and the WHO have summarised some of the perceived strengths of SHI-based versus tax-based funding models, primarily to inform debates about developing health care funding systems in middle- and lower-income countries (Box 9).,

Box 9: Perceived strengths of SHI-based versus tax-based funding models

The perceived strengths of SHI-based models are as follows:

  • SHI systems are more economically stable, as they avoid the unpredictability of tax revenues rising and falling as the economy changes.
  • SHI systems are more politically independent, because they avoid the risk that central government or finance ministries can reduce the flow of tax-based revenues to health ministries.
  • SHI systems tend to have purchasers (insurance bodies) separate from providers and this purchaser–provider split is a route to better value.

By contrast, the perceived strengths of tax-based systems have been characterised as follows:

  • Tax-based systems are simpler and less expensive to administer.
  • Tax-based systems are more progressive (as income is more likely to be raised in proportion to people’s ability to pay).

In the subsections that follow we look at selected recent developments in the seven countries under analysis in this report to explore the supposed strengths and weaknesses of each of the two models, under the three broad areas:

  • the experience of raising revenue and containing costs
  • ensuring sources of revenue are sustainable
  • the fairness of revenue raising, including user charges.

The experience of raising revenue and containing costs

Do social insurance systems spend more on health care?

There is no simple link between funding models and overall levels of spending on health care. The WHO’s 2009 analysis of 29 OECD countries found that SHI systems tended to be more expensive than tax-based systems by between 3% and 4%. Taking a snapshot of recent spending data (from 2021) from the OECD (and excluding the US as an outlier), there is a mix of tax-funded and SHI systems, if countries are ranked by spending per head (Table 10) or by proportion of GDP (Table 11).

Table 10: Health care spending for OECD countries, 2021, ranked by spending per head

Country

Government/compulsory spending per head, US dollars, 2021

Predominant financing type 

Germany 

6,424

SHI  

Norway

6,025

Tax

The Netherlands

5,722

SHI

Denmark

5,429

Tax 

Luxembourg

5,397

SHI 

Sweden

5,351

Tax

Austria

5,241

SHI

France

5,178

SHI

Switzerland

5,135

SHI

Belgium

4,674

SHI 

Note: The UK is 13th with US$4,539 per head.

Source: OECD, 2023, Health spending [indicator] (doi: 10.1787/8643de7e-en).

Table 11: Health care spending for OECD countries, 2021, ranked by spending as a percentage of GDP

Country

Government/compulsory spending as a percentage of GDP, 2021

Predominant financing type 

Germany 

11.1

SHI  

France

10.4

SHI

UK

10.3

Tax

Japan

9.7

SHI

Sweden

9.7

Tax

The Netherlands

9.6

SHI

Austria

9.5

SHI

Denmark

9.2

Tax

Canada

9.0

Tax

Belgium

8.6

SHI 

Source: OECD, 2023, Health spending [indicator] (doi: 10.1787/8643de7e-en).

Containing costs in social insurance systems: not so independent of politics?

One of the perceived attractions of SHI is a greater separation from central government, including more stability in funding as tax-based budget allocations are not potentially beholden to the ‘whims of policymakers’. In practice, central government plays an active role in both France and the Netherlands in setting and enforcing budgets. Germany’s government has also intervened to influence spending, revealing some of the complexities of managing a more devolved social insurance system.

In France, for many years, the norm was for the social insurance schemes to automatically reimburse services chosen by patients and provided by their doctors (often on a fee-for-service basis), and freedom of choice was (and still is) valued by patients. In 1996, faced with growing deficits in the social security budget, the Social Security Financing Act was passed to bring a much more active role for central government in controlling spending on health care (and other areas of social security spending). Parliament now votes annually on the maximum amount of growth in health spending for each sector, based on plans agreed between the government, social insurance bodies, and hospital providers. Despite the 1996 Act, for the next decade, targets were regularly exceeded and deficits grew. Further reforms were needed, including an early warning system to give alerts that targets were being exceeded, and the power to freeze or reduce payments for hospitals and other providers. Since 2010, the spending targets have been met, and the rate of annual growth reduced (from 4.9% in 2004 to 2.3% in 2018).

In the Netherlands, the market-inspired reforms of the health system in 2006 grew out of the repeated failures, from the 1970s, to control health care spending by setting annual hospital budgets or overall health spending caps. The creation of competition between health insurers via the 2006 reforms was designed to incentivise insurers to negotiate lower prices with providers, as well as drive up quality. Growth in overall health spending in the Netherlands began slowing from 2008, and was lower than in many other European countries between 2013 and 2018. There is evidence that, from 2012, insurers succeeded in negotiating lower prices for drugs, mental health care and GP care, although underspends in these sectors were compensating for overspends in the hospital sector.

What drove the success of these cost-containment efforts is still a matter of debate. Sharpened incentives for insurers played a role, as reforms brought in full liability for insurers for any deficits, and the removal of compensation for overspends. But so did government intervention through setting budget ceilings: maximum annual growth rates for different sectors are agreed through negotiations between the ministry of health, insurers, providers and patients.

Similar concerns about growing deficits arose in the 1990s in Germany, as sickness funds consistently spent more on care than the contributions they received. From 1993, the government attempted to impose legally binding caps on spending, and introduce more competition between the sickness funds, while still preserving the tradition of self-governing arrangements over planning and budgeting between sickness funds and providers. Reforms to hospital payment systems and the introduction of selective contracting were subsequently introduced in the 2000s to sharpen the incentives for hospitals to become more efficient.

The relative strength of Germany’s economy since then has reduced the pressure to contain costs and the complex, self-governing characteristics of Germany’s health care funding and planning have endured. Regional associations of sickness funds still hold separate budget negotiations with ambulatory services (local associations of GPs and non-hospital-based specialists) and with individual hospitals, reinforcing the silos between these sectors. The federal government negotiates the overall rules for pricing with national representatives of sickness funds and providers, but local budgets are largely agreed on the basis of the previous year’s spending. The result has been sustained growth in hospital activity, seen by some as potentially wasteful oversupply. In 2018, Germany had 19 million hospital discharges, the second-highest level in the EU (after Bulgaria), with limited levers for the federal government to directly intervene.

More scope for government intervention in tax-funded systems?

Social insurance systems do not, therefore, automatically mean insulation from government action, as can be seen in the case of France and the Netherlands. But tax-funded systems usually have more direct levers for influencing health spending, through the allocation of budgets by central government, and the scope for action is large.

This can be a weakness as much as a strength, most obviously so in the case of the UK. Health spending has gone through cycles of ‘feast or famine’ growth followed by retrenchment, which has resulted in a health system struggling to meet demand. Between 1997 and 2009, total health spending grew in real terms by 5.3% a year. This was followed by a period of austerity in which spending grew much more slowly, in real terms by 1.9% a year on average between 2009 and 2018. Another shift in government priorities in 2018 resulted in faster funding growth, a 3.4% a year average real-terms increase over 5 years, still lower than the historical average growth of 3.7%. These fluctuations in spending growth have contributed to inadequate investment in staff training, and capital investment in beds and equipment, which has left the UK vulnerable to external shocks, such as the COVID-19 pandemic.

The decentralised tax-based systems analysed in this report add a layer of complexity, because revenue is raised by both central and local government, and regions often have greater autonomy in decision making as a result. In the case of Spain, the decentralised system did not prevent nationally led reductions in health spending. Spain was hard hit by the financial crisis of 2008 and was required by the EU stability programme to reduce public spending on health from 6.5% of GDP in 2010 to 5.1% by 2015. This was achieved by reductions in wages and numbers of staff (many of whom are salaried employees in the NHS in the country) and changing the scope of user charges and who was entitled to health coverage.

An OECD survey of government officials in 27 countries in 2015 found that decentralisation was seen as a mixed blessing. Officials from Sweden believed that decentralisation had made it easier to control costs (county councils and municipalities have been required to balance their budgets since 2000, with deficits to be eliminated within 3 years by reducing spending or raising local taxes). On the other hand, it could result in geographical inequalities in health spending (reported by Italy and Spain) and soften budget control, resulting in a ‘blame game’ between different levels of government, where ultimate responsibility for staying within budget is not clear.

In Italy, where regions have been responsible for the organisation and management of health services since 2001, the emergence of regional deficits from 2008 led to increasing levels of intervention by central government, via financial recovery plans to control budgets. One study has looked at the impact of these recovery plans in 10 (of 20) regions, mainly located in the poorer south of Italy. The study found that the recovery plans led to spending cuts of 3.8% a year on average, achieved by cutting staff and beds, and estimated that they had also resulted in a small increase in avoidable deaths.

Administrative costs: higher in social insurance systems?

SHI has been linked with higher administrative costs., In a 2017 study of waste in health care, the OECD noted that SHI systems spend more on administrative costs than tax-based systems, and that some cases, where there is a free choice of competing insurers (such as Germany), were more likely to be associated with higher administrative costs. Table 12 shows that administrative costs are higher in the three SHI systems included in this report.

Table 12: Share of public health expenditure on governance and administration (%), 2021

Country

Share of spend (%)

Germany

3.9

France

3.6

Netherlands

3.2

Spain

1.2

Sweden

1.0

UK

1.0

Italy

0.8

Source: OECD health data, 2021.

Insurance companies in Germany and the Netherlands spend money on marketing and advertising, as they are expected to compete. On the government side, regulators are needed to monitor the policies and premiums that insurers offer, as well as scrutinise mergers and takeovers. The negotiation and administration of contracts between payers and providers also take place in tax-based systems, but the complexity, and costs, of multiple insurers negotiating contracts with multiple providers are likely to be higher.

Purchaser–provider splits exist in both types of funding systems

A perceived advantage of social insurance systems has been the separation of insurance bodies (purchasers) from providers, particularly if the purchasers are competing, compared with a state-run system. In recent decades, tax-based systems have increasingly adopted purchaser–provider splits. This has happened most comprehensively in England since the 1990s, but also in the decentralised systems. In Italy, for example, the region of Lombardy has adopted a quasi-market system, offering patient choice and fostering competition between public and private hospital providers, in contrast to other regions with more centrally planned and publicly provided services. Similarly in Sweden, in the 1990s, several county councils including Stockholm set up separate purchasing organisations, brought in contracting and new payment mechanisms for providers, and in some cases contracted with more private providers.

The evidence on how effective these sorts of reforms (known as ‘strategic purchasing’) have been since the 1990s is mixed, and many countries have struggled to make strategic purchasing work. Of the countries assessed here, the Netherlands represents a comprehensive attempt to embed strategic purchasing by insurance companies, but there is still little evidence that it has been a decisive factor in reducing costs or improving quality. The equivalent reforms in England, in which purchasing is known as ‘commissioning’, have proved complex to implement and have not been associated with any improvements in outcomes.

Ensuring revenue sources are sustainable?

Social-insurance-based systems historically relied on revenues derived from employment, which requires a stable labour market. Concerns about the impact of wage-based contributions on the competitiveness of labour in a global market and, more recently, about the consequences of ageing populations supported by smaller working-age populations, have seen a trend among some social-insurance-based systems to diversify their revenue streams, and include more tax-based funds. This is true of both France and the Netherlands, and to a more limited degree Germany.

Between 1947 and 1998, the French health system was funded almost entirely from employment-based contributions, split between employer and employee. In 1991, a new earmarked tax on personal income was introduced, which was applied to income from employment, but also income from financial assets, investments, benefits and pensions (with caps for those on low incomes) and, most recently, income from gambling. This reflected a long-term change in the sources of household income over the previous 40 years, where the share of income in household finances that came from employment fell from 80% in 1970 to 71% in 2011, replaced by income from capital and benefits. In 2021, payroll contributions accounted for 33% of SHI revenues, and the earmarked tax accounted for 24%. France has also introduced a range of smaller earmarked taxes, including on tobacco, alcohol, health insurance and company cars, and companies with sales over a certain amount. These taxes, plus contributions from VAT, contributed 33% of SHI revenues in 2021.

In the Netherlands, general taxation is also an important component of revenue for health care, accounting for 13% of spending on health care, compared with 72% from compulsory contributions to insurance companies (which include the income-related contributions paid via employers and premiums paid direct to insurance companies). Tax-funded subsidies for people on low incomes to help with the costs of their insurance premiums were introduced with the reforms in 2006. Tax revenues are also used to pay for the care of people younger than 18 years and preventive services such as vaccinations and cancer screening.

Tax revenue has also played a greater role in funding Germany’s health care system since 2004. There were concerns in the early 2000s about the burden of insurance contributions on employers, as the economy experienced a period of stagnation and unemployment. Reforms were introduced to increase the proportion of contributions coming from employees versus employers (54% and 46% respectively) and to fix the employer contribution rate to make it predictable. Although these reforms were later reversed (and contributions are currently split evenly between employer and employee), sickness funds can impose an additional income-related premium on employees. The 2004 reforms also brought in tax-based federal subsidies to the insurance system to cover maternity benefits, in-vitro fertilisation and other family-based policies. The introduction of a tax-based revenue stream was controversial. An important principle of Germany’s SHI system was that contributions were kept separate from general taxation, and some have described contributions based on the federal budget as inherently unstable because of the political influence over spending decisions. In official spending accounts, these tax subsidies are not identified separately, but combined with social insurance expenditure. They are estimated to be around 10% of total health expenditure.

The addition of tax funding in Germany has been seen as a positive in one respect: counteracting (to some degree) the drawbacks of a dual private–public insurance system. A prerequisite of publicly funded health systems, whether tax-based or social insurance-based, is spreading contributions across the entire population, by ensuring that people are not able to opt out. Of the countries surveyed in this report, only Germany offers high-earning residents the possibility of opting out of the public system. Critics argue that the public system is deprived of the income that would otherwise have come from these high earners, and that the public system is forced to cover a pool of people with above-average risk of ill health, as wealthier people tend to be healthier. As everyone pays income tax, the tax subsidies offset this to some degree. But, the ‘two-tier’ system has proved difficult to end, as it derives powerful support from the medical profession (who are paid more for treating privately insured outpatients) and the wealthy section of the population who benefit from it.

Are tax-based systems fairer than social insurance systems?

A ‘fair’ or equitable health system has two main components: it should raise funds in proportion to people’s ability to pay and distribute health care according to need, not ability to pay. To understand the former – whether revenue raising is equitable – all streams of revenue need to be taken into account, including taxes, social insurance contributions, private insurance and out-of-pocket payments. Studies of OECD countries based on data from the 1980s and 1990s found that tax-based systems in general tended to be more progressive (that is, raising money in proportion to people’s ability to pay) than social insurance-based systems. But just how much each country was progressive or regressive varied according to how policies were designed, for example whether people on a low income or pensioners were mostly exempted from social insurance contributions, or to the mix of taxes (indirect taxes, such as VAT, tend to be regressive). In all systems, out-of-pocket payments were likely to be ‘highly regressive’ unless exemptions were put in place.

The fairness of social insurance systems depends on the degree to which contributions are linked to people’s ability to pay. In France and Germany, social insurance contributions are income related. In the Netherlands, half of insurance revenues are income related, but the remainder come from flat-rate premiums paid directly to the insurers. To protect low-income families from the cost of these premiums (which rose sharply between 2005 and 2006 after the reforms were brought in), a tax-financed subsidy was introduced (known as the ‘care allowance’), which was claimed by 30% of the population in 2019. The subsidy has been controversial: critics view it as an administratively wasteful mechanism to move large amounts of tax-based money around the system.

Tax-based systems are not automatically progressive and much depends on the mix of taxes that are used. A recent study from Italy analysed the impact of decentralising the health system in the late 1990s, which allowed regions to vary how much people were charged for services, among other reforms. The researchers found that Italy’s financing system as a whole was regressive (due to increasing reliance on VAT as a source of revenue for health care) and that some regions, particularly in the poorer south, were more regressive than the north.

Out-of-pocket spending

Another method of assessing the equity of financing is by looking at the proportion of revenue that is out of pocket, which is considered to be the most regressive form of raising revenue. Comparable data are collected on out-of-pocket spending, which includes payments for medicines and medical supplies, consultations, diagnostic tests, hospital stays and complementary or alternative medicines. The OECD publishes these costs as a percentage of total health expenditure, and there is considerable variation between countries (Table 13).

Table 13: Household out-of-pocket spending as a percentage of total health expenditure (%), 2021

Country

Household out-of-pocket spending (%)

Italy

21.9

Spain

21.0

Sweden

13.1

UK

12.7

Germany

12.0

The Netherlands

9.4

France

8.9

Source: OECD Health Statistics. Health expenditure and financing.

The tax-based systems, particularly Italy and Spain, report higher levels of out-of-pocket spending than the three social insurance-based systems. In Italy, there has been a growing number of people willing to pay directly for faster access to health care from private providers, particularly for specialist consultations and diagnostic tests. In 2019, 26% of Italy’s total health care spending came from private sources, and out-of-pocket spending made up the majority of this (89%), spent on dental care, outpatient consultations and prescription drugs. Private health insurance is relatively small (accounting for just over 2% of total health expenditure). In 2019, 44% of Italians paid directly for at least one health care service without attempting to use the public system, with a higher proportion (50%) of those with a high income doing so, compared with 38% of those on a low income.

A full understanding of the degree to which these costs are equitable or not requires analysis of their impact on household spending. The WHO’s Regional Office for Europe has analysed national household survey data across Europe. In Spain, for example, despite higher levels of out-of-pocket spending compared with other countries, the incidence of ‘catastrophic’ spending (defined as exceeding 40% of a household’s capacity to pay) was relatively low (Table 14). In Spain, 1.6% of households experienced catastrophic spending in 2019, concentrated in the poorest quintile, driven mostly by dental and outpatient medicine costs. In Italy, by contrast, it was 9.4% in the same year.

Table 14: Share of households with catastrophic spending on health (%)

Country

Share of households, % (year)

Italy

9.4 (2019)

Germany

2.4 (2018)

France

2.1 (2017)

UK

1.5 (2019)

Sweden

1.7 (2012)

Spain

1.6 (2019)

Note: Data for the Netherlands (0.5%, 2015) are not directly comparable to other countries as the household survey does not include the out-of-pocket payment of the deductible.

Source: WHO Barcelona Office for Health Systems Financing.

User charges are common in both systems

One of the components of out-of-pocket costs is user charges for statutory health care. The presence or extent of user charges has no connection with funding models. Both the tax-funded and social insurance countries examined in this report raise some revenue from user charges, including the UK, where charges apply to dental care across all four nations of the UK, and to prescription drugs in England (Table 15).

Table 15: User charges by country

Source: Authors’ analysis.

User charges are often designed to have two aims: to generate additional revenue and to dampen demand for health care. Not all countries publish data on the amounts raised. Where they do, for example in Germany, the funding raised is found to be only a modest proportion of the total amount spent on health care (1.1% in 2019). Evidence on the effectiveness of user charges in dampening demand is limited. Experimental evidence from the US suggests that user charges reduce both necessary and unnecessary demand, and that the poorest and sickest patients suffer most as a result. All the countries analysed in this report have developed policies to protect the most vulnerable, and these policies have often required revision and adjustment.

In 2012, the government in Spain reformed user charges for outpatient prescription drugs, as part of a series of policies to reduce spending in the wake of the financial crisis. Exemptions for pensioners were abolished and charges increased for everyone else. Some financial protection was included, for example caps on the maximum amount that pensioners had to pay, which varied by income. Researchers found that there was a short-term reduction (of 18 months) in the number of prescriptions for drugs to treat chronic illnesses, including anti-diabetic drugs, and that some vulnerable people reduced their use of medicines (for example, drugs prescribed following a heart attack). The incidence of catastrophic spending rose from 0.6% of households in 2006 to over 2.0% in 2014, before falling back to 1.6% in 2019, concentrated in the poorest fifth of the population. Spending on dental care (minimally covered by the Spanish NHS) drove most of this spending, but the share of costs from outpatient medicines has steadily increased since 2012. In 2021, as concern grew about the impact of charges, heightened by the COVID-19 pandemic, exemptions from outpatient medicine charges were extended to low-income pensioners, moderately and severely disabled children, and households in receipt of child benefits.

In Germany, a new user charge for outpatient visits was abandoned within a decade. The reforms introduced in 2004 imposed a new charge for visiting an ‘ambulatory care’ doctor (office-based specialist or GP) of 10 euros per quarter, and for each contact thereafter with a physician without a referral. (At the same time, charges for outpatient medicines were changed from a fixed co-payment to a percentage co-payment.) Early evaluations of the new outpatient charge found that visits to doctors fell in 2004 compared with 2003, with no evidence of a drop in necessary visits by disabled people and people with chronic conditions. But patient–doctor contacts rose in subsequent years, suggesting that the effect on reducing demand was temporary. The charge was also unpopular with patients and the medical profession, and brought an additional administrative burden. It was abolished in 2012.

The outpatient charge had a tangible impact on household spending between 2003 and 2013. The share out-of-payment pocket payments rose in all income groups between 2003 and 2008 before decreasing in 2013 (after the abolition of the outpatient charge). The rate of increase was steepest among those in the two poorest quintiles. The outpatient share of out-of-pocket costs increased fourfold for those in the poorest quintile (from 4% to 18%) between 2003 and 2008, before dropping in 2013.

All systems that impose user charges have policies designed to financially protect those people least able to afford payments. These bring additional administrative costs and can also create barriers for claimants. In France, the majority of people buy supplementary private insurance to offset the user charges imposed on a wide range of health services (see Table 15). Until 2019, people on low incomes and those just above the poverty line were protected from charges through two schemes to provide access to insurance, financed by a tax on insurance companies. One of these, a voucher scheme, had only been taken up by a quarter of those potentially eligible in 2015, and research suggested that the administrative burden of applying was a barrier. The schemes have since been merged, but in 2019, 5% of the French population did not have private insurance. A survey of more than 150,000 people in 2019 found that a quarter had forgone health care in the previous 12 months, and nearly 60% reported that the charges were too high, even if people had additional insurance.

In Sweden, although overall catastrophic spending is low (1.7% of households in 2012), the WHO has calculated that it is highly concentrated in the poorest quintile, affecting about 6% of households, driven by the costs of outpatient medicines. The WHO concluded that the system for protecting poorer households from the burden of co-payments is bureaucratic and could be improved: people in receipt of social benefits have to apply to their municipality for retrospective reimbursement of health care charges, or request an invoice from the region (which organises health care) to be sent to the municipality to pay on their behalf.

Previous Next