Annex Table 6 reports adjusted scores for overall responsiveness, as well as a measure of
fairness based on the informants’ views as to which groups are most often discriminated
against in a country’s population and on how large those groups are. Either a larger group
being affected, or more informants agreeing on that group’s being treated worse than some
others, implies more inequality of responsiveness and therefore less achievement of fairness.
Since some elements of responsiveness are costly, it is not surprising that most of the
highest ranked countries spend relatively large amounts on health. They are also often
countries where a large share of provision is private, even if much of the financing for it is
public or publicly mandated. However, the association with a country’s income or health
expenditure is less marked than it is for health status. Several poor African and Asian countries
rank fairly high on the level of responsiveness. And countries that perform well on
average for responding to people’s expectations may nonetheless rank much lower on the
distributional index.
Fair financing in health systems means that the risks each household faces due to the
costs of the health system are distributed according to ability to pay rather than to the risk
of illness: a fairly financed system ensures financial protection for everyone. A health system
in which individuals or households are sometimes forced into poverty through their
purchase of needed care, or forced to do without it because of the cost, is unfair. This situation
characterizes most poorer countries and some middle and high income ones, in which
at least part of the population is inadequately protected from financial risks (12).
Paying for health care can be unfair in two different ways. It can expose families to large
unexpected expenses, that is, costs that could not be foreseen and have to be paid out of
pocket at the moment of utilization of services rather than being covered by some kind of
prepayment. Or it can impose regressive payments, in which those least able to contribute
pay proportionately more than the better-off. The first problem is solved by minimizing the
share of out-of-pocket financing of the system, so as to rely as fully as possible on more
predictable prepayment that is unrelated to illness or utilization. The second is solved by
assuring that each form of prepayment – through taxes of all kinds, social insurance, or
voluntary insurance – is progressive or at least neutral with respect to income, being related
to capacity to pay rather than to health risk.
Out-of-pocket payments are generally regressive but they can, in principle, be neutral
or progressive. When this happens, and out-of-pocket expenses are not too large, they
need not impoverish anyone or deter the poor from obtaining care. However, of all the
forms of financing they are the most difficult to make progressive. Arrangements that exempt
the destitute from user fees at public facilities, or impose a sliding scale based on
socioeconomic characteristics, are attempts to reduce the risk associated with out-of-pocket
payments (13, 14). Except when private practitioners know their clientele well enough to
discriminate among them in fees – and the better-off accept that their charges will subsidize
the worse-off – such arrangements are limited to public facilities, which often account
for only a small share of utilization in poor countries. And even then, such schemes require
relatively high administrative costs to distinguish among users, and typically affect only a
small amount of total risk-related payments.
For this reason, financial fairness is best served by more, as well as by more progressive,
prepayment in place of out-of-pocket expenditure. And the latter should be small not only
in the aggregate, but relative to households’ ability to pay. Prepayment that is closely related
to ex ante risk, as judged from observable characteristics – risk-related insurance premiums,for example – is still preferable to out-of-pocket payment because it is more predictable,
and may be justified to the extent that the risks are under a person’s control. However, the
ideal is largely to disconnect a household’s financial contribution to the health system from
its health risks, and separate it almost entirely from the use of needed services. The question
of how far insurance prepayments may be related to risks, and how such premiums
should be financed, including subsidies for those unable to pay, is treated in Chapter 5.
Ex post, the burden of health financing on a particular household is the share that its
actual health expenses are of its capacity to pay. The numerator includes all costs attributable
to the household, including those it is not even aware of paying, such as the share of
sales or value-added taxes it pays on consumption, which governments then devote to
health, and the contribution via insurance provided, and partly financed, by employers.
The denominator is a measure of the household’s capacity to pay. In poor households, a
large share goes for basic necessities, particularly food, whereas richer households have
more margin for other spending, including spending on health care. Food spending is treated
as an approximation to expenditure on basic needs. Total non-food spending is taken as an
approximation of the household’s discretionary and relatively permanent income, which is
less volatile than recorded income (15) and a better measure of what a household can
afford to spend on health and other non-food needs.
In sum, the way health care is financed is perfectly fair if the ratio of total health contribution
to total non-food spending is identical for all households, independently of their income, their
health status or their use of the health system. This indicator expresses the trenchant view of
Aneurin Bevan, that “The essence of a satisfactory health service is that the rich and the
poor are treated alike, that poverty is not a disability, and wealth is not advantaged.” (16).
Clearly the financing would be unfair if poor households spent a larger share than rich
ones, either because they were less protected by prepayment systems and so had to pay
relatively more out of pocket, or because the prepayment arrangements were regressive.
But to identify fairness with equality means that the system is also regarded as unfair if rich
households pay more, as a share of their capacity. Simply by paying the same fraction as
poor households, they would be subsidizing those with lower capacity to pay. It is true that
well-off households might choose to pay still more, particularly by buying more insurance,
but that can be considered equitable only if the extra spending is prepaid and if the choice
is entirely voluntary and not determined by the system of taxes or mandatory insurance
contributions.
Families that spend 50% or more of their non-food expenditure on health are likely to
be impoverished as a result. Detailed household surveys show that in Brazil, Bulgaria, Jamaica,
Kyrgyzstan, Mexico, Nepal, Nicaragua, Paraguay, Peru, the Russian Federation, Viet
Nam and Zambia more than 1% of all households had to spend on health half or more of
their full monthly capacity to pay, which means that in large countries millions of families
are at risk of impoverishment. Invariably the reason is high out-of-pocket spending. This
high potential for financial catastrophe has much to do with how the health system is
financed, and not only with the overall level of spending or the income of the country.
The fairness of the distribution of financial contribution is summarized in an index which
is inversely related to the inequality in the distribution, and presented in Annex Table 7. The
index runs from zero (extreme inequality) to 1 (perfect equality). For most countries, and
particularly for most high income countries, the value is not far from 1, but great inequality
characterizes a few countries in which nearly all health spending is out-of-pocket, notably
China, Nepal and Viet Nam. However, in some countries where most spending is out-of pocket, there is nonetheless little inequality because that spending is relatively progressive
and few families spend as much as half their non-food expenditure on health. Bangladesh
and India are examples. Generally, high values of equality are associated with predominantly
prepaid financing, but Brazil shows extreme inequality despite a high share of prepayment,
because of the great inequality in incomes and the large number of families at
risk of impoverishment.
The summary measure of fairness does not distinguish poor from rich households. Figure
2.5 introduces this distinction, by showing how the burden is distributed across deciles
of capacity to pay, and divided between prepayment and out-of-pocket spending, in eight
low and middle income countries. Prepayment is clearly progressive – the rich contribute a
larger share – in Mexico and the United Republic of Tanzania, and also in Bangladesh and
Colombia (not shown). It is actually regressive in India and Pakistan, and also in Guyana,
Kyrgyzstan, Nepal, Peru and the Russian Federation (not shown). In other countries –
Brazil, Bulgaria, Jamaica, Nicaragua, Paraguay, Romania and Zambia – the prepaid contribution
is distributed more or less neutrally or varies irregularly. Out-of-pocket spending
shows more variation, as might be expected; for example, it is progressive in India and quite
regressive in Pakistan and Viet Nam, where there is almost no prepaid financing at all.
Total non-food spending also includes whatever the household spends out of pocket on
health care. That spending is largely unpredictable or transitory, so to include it may overstate
the family’s capacity to pay. If out-of-pocket expenditure is small, it makes no difference;
but if it is large, it may have been financed by selling assets, going into debt, requiring
more family members to