It is well known by now that
population ageing jeopardises the sustainability of public finances in a number
of countries. The gradual retiring of the baby-boom generations, low fertility
rates and ongoing reductions in mortality rates portend dramatic changes in
the age structure of populations. In many countries, old-age dependency ratios
may more or less double within a period of 40 years. By themselves, these
changes would not be problematic, except for the pay-as-you-go (PAYG) nature
of many social security institutions. Population ageing unbalances the relation
between pension expenditures and pension contributions because of its PAYG
financing mode. This unbalance will be reflected in increasing fiscal deficits,
which cannot be expected to disappear if policies are left unchanged.
A number of studies address
the problem of quantifying the fiscal impact of population ageing. Especially
important are the studies by the EU and the OECD, which do so for a large
number of countries (Economic Policy Committee or EPC, 2001 and Dang et al.,
2001, respectively). Unfortunately, these studies rely on assumptions that
are difficult to accept in the light of recent empirical evidence. This also
holds true for the assumptions these studies make on mortalityrelated costs,
the future evolution of mortality rates and the health status of the population.
In particular, these studies neglect mortality-related costs, assume a slowing
down of the process of increasing longevity and postulate that the health
status of the population will, apart from the impact of ageing, remain unchanged.
Several arguments call into doubt the usefulness of these assumptions.
Three critical assumptions
The view that health-care expenditure is a function of age
alone is heavily debated nowadays. First, evidence abounds that health-care
expenditures by people in the last year of their lives is substantially larger
than those of survivors of the same age. Focusing on the last year of life,
the costs of decedents can be higher than those of survivors by a factor of
6. The share of expenditures during the last year of life of total health-care
spending on the elderly is more than a quarter. Furthermore, this share is
surprisingly stable over time. Calculations that neglect this type of evidence
produce estimates of expenditure growth that are way too high. The errors
involved may be 20% or higher. Second, many studies take the view that there
will be fewer gains in life expectancy because of biological limits. The idea
that life expectancy gains in the near future will be modest because life
expectancy is close to a biological limit has some intuitive appeal. Yet it
conflicts with more recent historical evidence. White (2002) concludes from
empirical evidence for a number of countries that life expectancy increases
by one year every five years. Over the last 40 years, the rate of growth in
life expectancy has not declined at all; it even shows a slight acceleration.
Furthermore, Vaupel (1998) presents
a number of historical examples in which the reductions in mortality rates
were highest for the oldest old, contrary to the argument of a biological
limit to life expectancy, which would suggest smaller life expectancy gains
for the older age cohorts. Similarly, according to the biological limit argument,
one would expect to observe smaller life expectancy gains for women, as on
average they live longer than men. Nevertheless, Kannisto et al. (1994) show
that in the 1980s, in contrast to the convergence argument, the gap between
the mortality rates of women and men did not decrease at all and even grew
further. To be sure, there is no reason to assume that the future is a mere
extrapolation of recent history. But it is also true that it is difficult
to consider a continuation of historical trends an unlikely scenario. A third
assumption that may be questioned concerns the health status of the population.
Most projection exercises that calculate the impact of changing age structures
assume constancy of the health status of the population per age group. Historical
evidence casts doubt on the validity of this assumption, however. Manton et
al. (1997), Jacobzone et al. (2000) and
Cutler (2001) document that disability rates among the elderly have declined
and that the health status of elderly persons has in general been improving.
Even if the more recent trend of worsening health as a result of overweight
and obesity continues, it is not to be expected that the historical trend
of improving health will halt within a few years time.
The future of fiscal deficits
and debt positions
The obvious question arises
of what will be the impact of alternative assumptions on these three aspects
for the future development of budget deficits. Health-care spending may be
seriously affected – not only the spending on acute health-care services,
but also the spending on longterm care services. Projections for pension expenditure
may be importantly altered as well. But the projections for labour market
participation and thus tax and social security revenues may also change on
account of alternative assumptions about the health development of the population.
Ultimately, alternative insights may then change our assessment of the fiscal
sustainability problem.
This report explores the impact
of alternative assumptions on the determinants of medical spending, the development
of life expectancy and the development of health. It covers the public sector
in a broad sense, i.e. it analyses health expenditures, pension expenditures,
social security expenditures and tax and social security revenues. It makes
calculations for the group of EU-15 countries. It assesses the impact of life
expectancy and health status separately and simultaneously, giving rise to
three alternative scenarios: ‘living longer’, ‘living in better health’ and
‘living longer in better health’. Yet a caveat is in order before presenting
the results. It would be tempting to interpret the calculations as projections
of the most likely future developments of important variables. We warn against
such an interpretation. The reason is that our calculations are kept deliberately
simple and omit several aspects that are important in real life in order to
focus on the contribution of the elements of mortality-related costs, life
expectancy and health improvements.
Our study is hopefully able to say something useful on the contribution of
these three variables but nothing on the contribution of all other variables
one can think of that will be relevant for fiscal sustainability projections.
In order to avoid any misunderstanding we do not present the base case scenario
but focus on the differences that relate to the trends in demography and health.
Our base case scenario does
however reflect some of the things we learned from earlier projection exercises.
During the next four decades, medical spending on acute health-care services
and long-term care services will increase, in absolute terms and as a percentage
of GDP. Pension expenditure will also increase, even faster than medical spending.
The increase in pension expenditure will peak somewhere around 2035, starting
to decline when the baby-boom generations gradually pass away. But the increase
in health expenditure will continue to reflect the ongoing increase in life
expectancy. This illustrates once again that population ageing is not a temporary
issue, which will be resolved once the baby-boom generations have disappeared.
The combination of an ongoing
increase in life expectancy with a fixed age of retirement implies a permanent
increase in the ratio of retirees to workers. We use the sustainability gap
to measure the size of the fiscal sustainability problem. To understand the
sustainability gap, note that population ageing implies a debt that does not
show up in official statistics. Summing the explicit debt and implicit debt
gives the total public debt. The sustainability gap is the annuity value of
this total public debt figure. We express the sustainability gap in terms
of GDP, as is usual for debt figures. Hence, the sustainability gap is the
immediate and permanent change in the primary surplus-to-GDP ratio required
to restore fiscal sustainability.
Why future prospects may be
brighter or duller
In terms of fiscal sustainability,
the impact of mortality-related costs is relatively modest. The sustainability
gap that corresponds to a scenario that does not take into account mortality-related
costs (and that is identical in all other respects to the base case scenario)
is only 0.2 percentage points higher than that of the base case scenario.
Despite its importance, health-care expenditure is only one of the budgetary
items affected by population ageing. Pension expenditure, social security
expenditure and taxes and social security revenues do not change when mortalityrelated
costs are included in the analysis. Focussing on health-care costs only, the
difference is about 15%, which is in line with a number of other studies that
simulate the impact of mortalityrelated costs for the future growth of health-care
spending. Compared with this, the impact of a stronger increase in life expectancy
is much larger. Our living longer scenario assumes an increase of eight years,
to be compared with a five-year increase in the base case scenario. Note that
this corresponds more closely with historical evidence, which has shown a
one-year increase in life expectancy every five years for a number of countries.
The sustainability gap for the EU-15-average is now 1.0 percentage points
of GDP larger than in the base case scenario, because the expansion of longevity
increases pension and health expenditure. It is noteworthy that the reduction
in mortality rates, which drives the increase in longevity, also reduces health
spending in a very direct way, namely by lowering mortality-related costs.
This effect is so small, however, that it is dominated by the boost in health
spending resulting from the expansion of longevity.
The impact of an alternative
assumption on the development of health is of similar importance. Assuming
an improvement of health, the sustainability gap falls by 0.8 percentage points
of GDP for the average EU-15 country. That the effect of a health improvement
is so large has to do with its multiple impacts. Better health not only reduces
health expenditure, but also delays retirement, thereby increasing participation
in the labour market and reducing social security expenditure.
Given that the impact of both
a stronger increase in life expectancy and a steady improvement in the health
of the population is relatively large, it is interesting to see the impact
on fiscal sustainability of the combination of these two trends. This effect
turns out to be rather small, however: the sustainability gap for the living
longer in better health scenario is almost similar to that in the base case
scenario. The drop in public spending related to healthier lives neutralises
the boost in public spending on account of longer lives. Yet on a lower aggregate
level, the combined scenario does not work out to be neutral. Pension expenditure
and expenditure on long-term care services increase faster than in the base
case, whereas acute health-care expenditure increases at a slower pace. Moreover,
the uncertainties are particularly large in the combined scenario.
A warning signal
The calculated sustainability
gaps deviate significantly from zero and the conclusion that current fiscal
policies in many EU-15 countries are unsustainable is pretty robust. Obviously,
exogenous developments may help to make the future look brighter. A substantial
increase in labour market participation would help to reduce fiscal sustainability
problems to a large extent, for example. In particular, if the future increase
in life expectancy is accompanied by a rise in the (actual) retirement age,
the extent of fiscal problems will decrease. On the other hand, there are
adverse risks as well. The prospects of an improvement in the health status
of the population may fail to materialise and health spending may increase
much faster than is assumed in our calculations. Indeed, there is ample evidence
that economic factors play an important role in health expenditure projections
and in the assessment of the sustainability of fiscal policies as well. Sustainability
gaps would then be much higher than those that follow from our calculations.
Assuming some risk aversion on the part of policy-makers, i.e. that they are
more concerned with the pessimistic scenarios than the more optimistic ones,
this only strengthens the case for policy reforms that help to close fiscal
sustainability gaps. Which policies should be reformed is a question that
we cannot answer and clearly falls beyond the scope of our analysis. What
our analysis offers is only a signal. The signal is that living longer in
better health will not relieve the fiscal sustainability problems in the EU-15
countries.
To read the report click
here
Accueil
