A quarterly magazine of the IMF

September 2006, Volume 43, Number 3

Can Europe Afford to Grow Old?
Giuseppe Carone and Declan Costello

The EU must face up to recent projections showing that aging will have a major economic and budgetary impact

The population of the 25-member European Union (EU) in coming decades is set to become slightly smaller—but much older—posing significant risks to potential economic growth and putting substantial upward pressure on public spending. The region's old-age dependency ratio (the number of people 65 and over relative to those between 15 and 64) is projected to double to 54 percent by 2050, meaning that the EU will move from having four persons of working age for every elderly citizen to only two. In addition, upward pressure on spending has fueled concerns that unsustainable public finances could jeopardize the smooth functioning of the single currency, the euro.

Population aging in Europe is occurring because of the interaction of four demographic developments. First, fertility rates in all EU countries are, and are projected to remain, below the natural population replacement rate. Second, the recent decline in fertility rates followed the postwar baby boom, and the impending retirement of these cohorts will lead to a transitory increase (albeit lasting several decades) in the old-age dependency ratio. Third, life expectancy at birth, having increased by eight years since 1960, is projected to rise by a further six years for males and five years for females by 2050, with most gains resulting from longer life spans. Fourth, large net migration inflows are projected up to 2050: although cumulating to close to 40 million people, they will not offset low fertility and growing life expectancy.

Indeed, according to official projections, between 2004 and 2050, the number of young persons in the EU (aged 0–14) will drop by 18 percent (see table). The working-age population (15–64) will fall by 48 million, or 16 percent, whereas the elderly population aged 65+ will rise sharply, by 58 million (or 77 percent), and the fastest-growing segment of the population will be the very old (aged 80+).

Governments have not been idle in the face of the aging challenge. At the European level, the European Commission has worked with national authorities in the Economic Policy Committee (EPC) to develop more comparable long-term projections to assess how population aging will affect EU labor markets, economic growth potential, and public finances. Published in early 2006, the projections—which assume no change in current policies and follow up on an assessment published in 2001—point to the need for substantial policy reform and adjustment.

Dramatic repercussions for growth and labor

Although most of the debate on aging concerns the budgetary cost, the most immediate impact will be felt in the labor market. EU projections, however, point to a mixed picture, with improvements in employment delaying somewhat the onset of the economic repercussions of aging.

For the EU as a whole, the employment rate is projected to rise from 63 percent of the potential labor force in 2004 to 67 percent by 2010 and to 70 percent by 2020 (see Chart 1). This improved performance results mainly from higher female employment rates, which are projected to rise from just over 55 percent in 2004 to almost 65 percent by 2025 and is explained by the gradual replacement of older women by younger women who have higher educational attainment and a stronger attachment to the labor market. In addition, the employment rate of older workers (ages 55 to 64) is projected to increase substantially, reversing the trend toward earlier retirement and reflecting the positive effects of recent pension reforms.

Rising employment rates, however, can provide only a temporary cushion; the weight of demographic change eventually prevails. Three distinct periods can be observed (see Chart 2). Until 2011, the size of the EU's working-age population and overall employment levels will continue to rise: this can be viewed as the tail end of the demographic dividend of the baby-boom generation where the bulge in the size of the working-age population is conducive to growth. Between 2012 and 2017, rising employment rates will offset the projected decline in the size of the working-age population brought about by the baby-boom generation entering retirement. After 2018, the trend toward higher female employment rates will have come to an end, and, in the absence of further pension reform, the employment rate of older workers is projected to reach a steady state. Having increased by some 20 million between 2004 and 2017, employment during this last period is projected to contract by almost 30 million, giving a reduction of 9 million over the entire 2004–50 time horizon. Taken as a whole, the labor force projection indicates that Europe has a fast-closing "window of opportunity," when conditions will be relatively favorable to enacting structural reforms.

For the EU as a whole, the annual average potential GDP growth rate is projected to decline from 2.4 percent during 2004–10 to only 1.2 percent between 2031 and 2050. For the 10 newly acceded members (EU10), the fall is much steeper, in part because of their less favorable demographic prospects. The sources of growth will alter dramatically. Employment will make a positive contribution to growth in both the core EU15 and the EU10 up to 2010, but becomes neutral during 2011–30 and turns significantly negative thereafter. Over time, labor productivity will become the dominant—in some countries, the only—source of growth (see Chart 3).

Significant projected increases in public spending

Coupled with the decline in the potential rate of GDP growth, the aging of the EU population is projected, on the basis of current policies, to lead to increases in public spending in most member states by 2050, although there are wide differences across countries (see Chart 4). For the EU as a whole, age-related spending in the baseline scenario is projected to increase by between 3 and 4 percentage points of GDP between 2004 and 2050, which would be equivalent to a 10 percent increase in the size of the government sector.

The bulk of the projected increase in public spending would go toward pensions, health care, and long-term care (see Chart 5). The potential savings on unemployment transfers and education are relatively small: it is arguable that in practice they may not materialize on education given pressures on governments to increase the quality of education and to develop life-long learning and training strategies to sustain productivity throughout longer working lives.

The projected change in public spending on pensions between 2004 and 2050 ranges from a decrease of 5.9 percentage points of GDP in Poland to an increase of 12.9 percentage points of GDP in Cyprus. Interestingly, very small increases in spending on pensions are projected in Sweden and (in the longer run) Italy because their schemes offer a notional rate of return tied to the growth in overall wage contributions, with individual benefits based on effective working-life contributions. Relatively moderate increases (between 1.5 and 3.5 percentage points of GDP) are projected in many other EU countries. Larger (more than 5 percentage points) increases are projected for Belgium, Cyprus, the Czech Republic, Hungary, Ireland, Luxembourg, Portugal, Slovenia, and Spain. The projected decreases in Estonia, Latvia, and Poland, as well as small projected increases in Lithuania and Slovakia, stem partly from pension reforms enacted during the past 10 years involving a partial switch from the public old-age pension scheme to privately funded schemes.

Regarding health care, the effect of an aging population alone would increase public spending by some 2 percentage points of GDP by 2050. But non-demographic factors also drive spending. For example, if healthy life expectancy evolves broadly in line with change in age-specific life expectancy, then the projected increase in spending on health care resulting from aging would be approximately halved. On the other hand, neither the effects of technology nor the prospects of demand for better-quality health care services have been modeled, and the projections may be significantly underestimating future pressures on spending coming from these sources.

Public spending on long-term care is even more sensitive to population aging than that on health care as disability rates rise very sharply with age, especially among those aged 80+. The EU rate is projected to rise by just under 1 percentage point of GDP by 2050 because of the aging of the population alone, but the increase varies across countries by between 0.1 percentage point of GDP in Poland and 2.4 percentage points in Sweden. This reflects hugely different approaches to the provision and financing of formal care. Countries with very small projected increases in public spending currently have limited formal care systems within the public sector. With an aging population, a widening gap may emerge between the number of elderly citizens with disabilities who need care and the supply of formal care services. This pressure for more formal care linked to aging may be further exacerbated by the reduced availability of informal care at home as more women participate in the labor market.

How should policies change?

So what are the main implications of these economic and budgetary projections? We would like to draw six main conclusions on how the policies of EU member states need to be adapted and also on how future common projections at the EU level could be improved.

First, the projections support the current thrust of the EU's economic policy strategy, as defined in the Lisbon agenda (laid out by the European Council in Lisbon in 2000). The strategy focuses on policies to raise labor utilization with an emphasis on older workers and extending working lives, as well as on measures to increase labor productivity. However, the projections also underline the dangers of the gap between commitments to undertake reform and concrete actions. Moreover, the window of opportunity during which employment and demographic trends are set to remain broadly favorable to reform is fast closing.

Second, pension reforms appear to be working. This conclusion is based on the recent upsurge in the employment rates of older workers in many European countries, plus a comparison of the recent pension projections with those made in 2001, which point to much lower projected increases in spending on pensions in countries that have instituted reforms, for example, Austria, France, and Germany. While some reforms may not be sufficient to cope with the economic and budgetary consequences of aging populations, real progress has been made in recent years. But there is an ever-present danger of backsliding as governments face constant pressure to enact special, more favorable pension or early retirement schemes for special interest groups.

Third, pension systems should be designed to be sustainable in the face of uncertain economic and demographic developments. Recent measures in countries such as France, Germany, Italy, and Sweden to link pension entitlement to future changes in life expectancy are an interesting step in this direction and can help prevent the need for frequent reforms to pension systems. Carone and others (2005) illustrate the dynamic nature of the challenge posed by continuous increases in life expectancy: although the EPC and European Commission (2006) project the average exit age from the labor market to increase by one year by 2050, life expectancy at retirement will increase by three years. Hence, even on the basis of reformed pension systems, additional reform to raise the average exit age by a further two years would be needed to stabilize the share of adult life (15+) spent in retirement at its 2004 level.

Fourth, aging may pose even more complex policy challenges in the area of health care and long-term care than pensions, especially when the effects of non-demographic drivers of spending, such as investment in medical research and in modern technologies, and the increasing demand for most advanced treatment are also considered. The control of public spending through aggregate cost-containment measures (controls on volume, prices, and wages, as well as budgetary caps) is likely to remain a key element of member states' comprehensive health care strategies: however, their effectiveness may diminish over time. Voters are growing older, making this older age group a constituency for higher spending that is difficult to resist. Suppliers may alter their behavior, and they risk introducing distortions that could lead to costly inefficiencies. Measures to improve cost efficiency, create incentive structures that encourage rational resource use, and achieve an effective management of technological advancement will be of critical importance.

Fifth, and of a more methodological nature, uncertainty in demographic and budgetary projections should be addressed more systematically. Like other organizations that make long-run forecasts, the EPC and the European Commission deal with uncertainty via deterministic sensitivity tests. But these provide no indication of the probability of future outcomes or confidence intervals based on past forecast errors. Despite the lack of consensus among demographers and economists on how best to deal with uncertainty (for example, through the use of stochastic projections), the issue warrants attention in future common projection exercises.

Finally, even though full data comparability has not been achieved, long-run cross-country projections can provide a valuable tool to facilitate the conduct of difficult policy debates. Nonetheless, efforts should be made in future projection exercises to improve comparability and transparency. Regarding comparability, particular attention could be paid to improving the underlying data on migration flows and on health and long-term care spending. For transparency, more information could be made available on the different models used by national authorities to project pension spending.

* * * * *

In sum, the aging of populations will have large repercussions for EU labor markets, economic growth, and public finances. Aging, however, is not a tsunami that will overwhelm public finances when the baby-boom generation starts to retire. It is a slow-moving, largely predictable process that is therefore manageable, provided policymakers act in an efficient and timely manner.