The fiscal cost of aging: the case of Belgium Open Access

Belgium faces a fiscal consolidation challenge at a time when the fiscal cost of aging is mounting. This article focuses on the direct fiscal cost of aging, but fiscal costs may also be indirectly affected by the macroeconomic impact of aging. Potential growth could be affected if aging results in reduced labor contribution and productivity growth. Because consumption changes over the life cycle, the structure of aggregate consumption is also likely to evolve. This would affect tax revenue and lead to changes in sectoral production and employment. For example, the OECD (2023) estimates that the share of long-term care workers in total employment would increase by 0.3 percentage point (ppt) between 2023 and 2033 and by another 0.5 ppt between 2033 and 2043. This would affect economy-wide productivity growth as “labor productivity growth in the Long-Term Care (LTC) sector is projected to be lower than in the overall economy.” Moreover, growth may be affected if the rising fiscal cost of aging crowds out more productive public spending.

This article is organized as follows. The first section documents the expected aging of the Belgian population. The second section quantifies the expected cost of aging and its underlying drivers. The third section focuses on pensions, showing that the faster increase in pension costs than in European peers is likely to continue if no measures are taken. It discusses policy options. The last section turns to healthcare, arguing that health outcomes are comparatively favorable in Belgium but achieved at a higher cost than in European peers. The section presents policies and reforms that could increase spending efficiency. Efficiency gains would help absorb part of the fiscal cost of aging related to healthcare. However, there is still a need to reorganize the healthcare system to prepare for the needs of an aging population.

The Belgian population is aging rapidly. Over 2005-2022, the number of Belgians 65 and older increased almost three times faster than the population below 65. Moreover, aging has been accelerating, while the population 65 and older grew less than the population 65 and younger during 2005-11, it grew nine times faster during 2016-2022. As a result, almost one Belgian out of five is now 65 or older (Table 1 and Figure 1).

Due to differences in assumptions, databases, starting points, and models, demographic projections differ significantly, although all concur that aging will be rapid (Box 1 and Figure 1). According to Europop 2019, the population will start declining in 2048, due to a decline in the population below 65 years. The share of the population 65 and older is expected to increase by over 8 ppts to 28 percent in 2070, driven by a 6 ppts increase in the population 80 and older. The updated Europop 2023 projects an increase in the total population throughout the period (Eurostat, 2023). The decline in the population below 65 is much smaller than previously projected, while the increase in the share of population 65 and over remains broadly unchanged. A third projection, conducted by the national authorities, projects a much faster increase in total population (9.2 percent between 2022 and 2070 versus 3.6 percent for Europop 2023). In contrast with Europop, the population below 65 would increase. The share of population 65 and over would increase less than projected in Europop 2023 (6.5 ppts versus 8.5 ppts).

Despite uncertainty in the pace of aging, various projections concur on important drivers of the fiscal cost of aging:

• •

  Old-age dependency will increase markedly. Europop and national projections result in a sharp increase in old-age dependency (Figure 2). This will have a strong impact on pension costs.

• •

  The elderly population is aging. The share of population that is 80 years and older, which already increased from 3.5 percent in 2000 to 5.5 percent in 2022, is projected to double by 2070. The population 80 and older accounted for 28 percent of the population 65 and older in 2022, and this share would rise to 40-42 percent in 2070 (Figure 3). The increase in aging intensity will have a strong impact on healthcare costs as health needs grow more than proportionally with age (Figure 4). Kotschy and Bloom (2022) estimate that the number of Belgian older than 65 with at least two limitations in Activities of Daily Living (ADL) or Instrumental Activities of Daily Living (IADL) will increase from 3 percent of population to 4.5 percent of population between 2020 and 2040. Care needs are measured by the need to be (1) assisted with personal care through help with activities of daily living (ADL) such as eating, washing, and dressing, and (2) assisted to live independently through help with instrumental activities of daily living (IADL) such as cooking, shopping, and managing finances.

With health and pension spending mostly publicly financed in Belgium, population aging poses a significant fiscal challenge (Figure 5):

Aging could increase fiscal spending by 5.4 percent of GDP between 2019 and 2070. Under the reference scenario of the 2021 Ageing Report (EC, 2021) and the 2023 report of the Belgian Study Committee on Aging, about half of the increase would be due to pensions and the other half to healthcare (including long-term care). However, uncertainty is significant: the Ageing Report’s alternative scenarios suggest that the projected increase in the fiscal cost of healthcare vary by ± 2 ppts of GDP (Figure 6).

A large share of the fiscal cost of aging will take place in the current decade. Therefore, the cost of aging will conflict with the need for sustained fiscal consolidation in the coming years. Between two-thirds (Study Committee on Aging) and two-fifths (Ageing Report) of the fiscal cost of aging for 2019-2070 is expected to take place by 2030 (Figure 7). The difference in the estimated increase in spending by 2030 is due to healthcare. The Study Committee on Aging projects that healthcare costs would increase faster than both the Ageing Report’s baseline scenario and adverse health risk scenario. The health risk scenario captures the impact of non-demographic factors on health and long-term care expenditure. Notably, this scenario assumes a partial continuation of upward healthcare costs and a stronger increase in the costs and coverage of long-term care. According to the Study Committee on Aging, 1.2 ppts of the projected increase in the cost of aging between 2019-70 had already materialized by the end of 2022. Another important difference between the two projections is that the healthcare costs of aging are driven by long-term care according to the 2021 Ageing Report, while it is driven by acute care according to the Study Committee on Aging. This difference has large implications: (i) the reforms and policies needed to prepare the healthcare system for long-term versus acute care needs are different, and (ii) the fiscal burden falls on different levels of government, as regions are in charge of long-term care, while acute care is financed by the federal government.

The cost of pensions has increased rapidly. From 2008 to 2020, the pension-to-GDP ratio increased by 23 percent. As this increase was more rapid than in EU peers, spending on pensions rose from below EU level to the EU average of 13.6 percent of GDP (Figure 8).

The strong increase in spending on pensions results from a rapid increase in beneficiaries. The number of pension beneficiaries increased by 31 percent in Belgium between 2008 and 2020, while it increased by 4 percent in the EU as a whole. In other words, the share of pension beneficiaries in the total population increased by almost 5 ppts, while it increased by 1 ppt in the EU (Figure 9).

By design, aging leads to an increase in people receiving old age and survivors' pensions. This type of pension is the primary driver of the increase in pension beneficiaries in the EU. In contrast, in Belgium, beneficiaries of old age and survivors’ pensions account for less than half of the total increase in pensioners. The main driver is the increase in beneficiaries of disability pensions. The measures taken in the 2010s (IMF, 2019; Deroose et al., 2023) to tighten early retirement eligibility led to a sharp decline in early retirement spending. However, this contributed to an increase in spending on sickness and disability (Figures 10 and 11).

Another reason for the rapid increase in pension spending is that, in contrast to EU peers, Belgium has not reduced the generosity of its pension system. Following Lindert (2021), generosity of the pension system is measured by the support ratio, defined as pension benefit per person over 65 divided by GDP per person of working age (18-64). Due to aging, the decline in the support ratio needs to be large to result in a decline in pension-to-GDP spending. In the recent past, through pension reforms, all peers reduced their support ratios, more than reversing the increase experienced during the Global Financial Crisis (GFC). Despite a rebound during the pandemic, their support ratios remain below pre-GFC levels. This allowed them to contain pension spending and, in some cases, to reduce it. In contrast, Belgium has not implemented sufficient measures to significantly reduce the generosity of pensions. The support ratio, which was below all peers up to 2008, is now the highest among peers (except France), and the pension-to-GDP ratio has since grown steadily (Figure 12).

Looking ahead, the 2021 Ageing Report projects that Belgium’s pension spending would increase by about 3 percent of GDP between 2019 and 2070, compared to an increase limited to 0.1 percent of GDP for the EU (Table 2 and Figure 7). The pension-to-GDP ratio would increase from the EU average level in 2020 to the third-highest in the EU in 2070 (Figure 13).

Aging is a key determinant of the projected growth in pension spending. The sharp increase in the dependency ratio (Figure 2) would increase pension spending by about 7 percent of GDP (Table 2). This is comparatively large. However, the case of France highlights that policy measures have a role to play. France faces a similar increase in the dependency ratio as Belgium, but its pension-to-GDP ratio is projected to decline, thanks to stronger containment from the coverage and benefit ratios (Table 2). France has implemented several pension reforms since 1993, which raised both the pensionable age and the contribution period necessary to reach full pension and eliminated most early retirement schemes. The impact of the pension reform adopted in 2023 is not incorporated in Figure 12.

Containing the rise of pension spending is necessary to ensure the sustainability of the pension system, avoid crowding out productive investment, and support fiscal consolidation. The drivers of the projected increase in pension spending identified by the Ageing Report (Table 2) highlight policy levers that can be used:

• a)

  Increase labor force participation. The Ageing Report projects that the change in employment rate will account for only one-third of the small labor market effect reported in Table 2. Measures to increase the labor force participation of older workers and incentivize them to work longer could have an important impact on spending by reducing the coverage ratio. The coverage ratio is the ratio of the number of pensioners to the number of people 65 and older. According to the National Bank of Belgium (NBB), if the authorities meet their objective to increase the relatively low employment rate to 80 percent, pension costs will decrease rapidly and strongly (Table 3).

• b)

  Tighten the eligibility for pensions. Measures to tighten the eligibility for pensions (e.g., eligibility for disability pensions) or further increase the retirement age can boost the relatively high contribution of the coverage ratio to the containment of pension costs (Table 2). The NBB estimates that reducing the coverage ratio in Belgium to the euro area average would reduce the pension costs by 1.1 percent of GDP in 2030 (Table 3).

• c)

  Avoid an increase in the generosity of pensions. The gross replacement rate is projected to increase in the current decade and to remain above its 2019 level until the early 2050s (Figure 14). The gross replacement rate is the average first pension as a share of the economy-wide average wage at retirement. Avoiding such an increase would help contain pension spending, especially if combined with other measures to accelerate the reduction in the benefit ratio, that is projected to be backloaded (Figure 15). The decline in the benefit ratio will be smaller than projected by the Ageing Report due to recent measures, such as the increase in the minimum pension and the increase in the upper wage limit used to calculate pension entitlements (Deroose et al., 2023). The benefit ratio is the average pension as a share of average compensation.

What would it take to stabilize the pension-to-GDP ratio by 2030 or by 2050? Preventing an increase in pension spending in the short-term would help implement the needed fiscal consolidation. Table 4 presents the results of macro-simulations of measures that could stabilize the pension-to-GDP ratio at its (high) 2020 level. These results present trade-offs between various policies and suggest that a combination of measures may be the simplest way to contain pension costs. They should be seen as illustrative because the model, though calibrated for Belgium, does not reflect all the specificities of the Belgian pension system.

Another set of macro-simulations illustrates how further increasing the retirement age and/or cutting benefits would contain spending over time. The first scenario simulates the impact of linking retirement age to life expectancy gains starting in 2030, when the current measure to progressively increase the retirement age for men and women from 65 to 67 will be completed. This would allow Belgium to reverse the increase in pension spending starting in 2035. The pension-to-GDP ratio would come back to close to its 2025 level by 2070 and be about 2 percentage points lower than in the baseline. This would require setting the retirement age to 70.5 years in 2070. The second scenario simulates the impact of a 10 percent reduction in benefits between 2025 and 2034 (1 percent each year) for all pensioners or only for new pensioners. The savings in the short-term are larger than under the retirement age scenario but, in the long-term, result in smaller savings (Figure 16).

In sum, due to aging, spending on pensions will continue to increase significantly and rapidly, requiring additional policy measures. The projected increase in pension spending is larger than for EU peers. Moreover, the EC and the national projections concur that pension spending during the current decade will increase by 1.75 percent of GDP. Therefore, additional spending containment measures are needed to prevent the increase in pension spending crowds out productive public spending and hampering fiscal consolidation.

Against most metrics, health outcomes compare well to EU averages (Figure 17). Moreover, self-reported unmet needs are lower than for the EU as a whole, although higher than in Germany and the Netherlands (Figure 18), and self-perception of health is, for all income quintiles, better than for the EU, France, Germany, or the Netherlands.

However, health spending is relatively high. Whether considered as a share of GDP or per capita, Belgium’s public health spending is significantly higher than the EU, OECD, or advanced economies’ averages. It is also higher than in large neighboring EU countries, except France (Figures 19 and 20).

Aging is putting health spending under pressure: people 65 years and over represent one-fifth of the population but account for half of healthcare expenses. Moreover, though elderly people tend to suffer from multimorbidity and people with chronic (multi)morbidity represent 29 percent of the population, they account for 65 percent of healthcare expenses (Van der Heyden et al., 2023).

Potential efficiency gains appear large. Belgium achieves a healthy life expectancy that is only marginally higher than that of the EU or the OECD, but at a substantially higher spending per capita (Figure 21). If Belgium's healthcare system were as efficient as those of the EU and the OECD averages, it could achieve the same outcome 20 to 30 percent more cheaply (Table 5).

A combination of measures can help increase efficiency. Based on Belgian (KCE, INAMI-RIZIV, Sciensano) and international studies (EC, OECD, WHO) as well as academic literature, four types of measures are likely to have the largest impact on efficiency: (1) increasing prevention, (2) fostering a more adequate use of resources, (3) reducing the cost of pharmaceuticals, and (4) reducing reliance on hospitals. In some areas, policies are being implemented, but they need to be scaled up (Gerkens and Merkur, 2020).

• 1)

  Prevention

Belgium’s healthcare system is better at treating diseases than at preventing them. The rate of preventable deaths is higher than in other EU countries with a similar level of health expenditure (Figure 22). Another illustration of the effectiveness of the healthcare system in curing diseases is the case of cancers which is important in the context of aging, as for many cancer types, the risk of developing the disease rises with age. Belgium has a higher incidence of cancer than the EU average but a lower death rate from cancer.

Spending on prevention is low. At 1.6 percent of all spending in 2019, spending on prevention is lower than the EU average of 2.9 percent. Only €61 per capita is dedicated to prevention which is 40 percent lower than the EU average of €102 per capita (OECD Observatory on Health Systems and Policies, 2021). Therefore, INAMI-RIZIV (2022) sees prevention as one of the “highly problematic areas” of the Belgian healthcare system, flagging “alarming indicators.”

Increasing spending on prevention would not only improve the population’s well-being but would also be cost-effective. Prevention should focus on risk factors that have a significant impact on health, such as obesity (De Pauw et al., 2022; Gerkens and Merkur, 2020). The fight against overweight and obesity would be cost-effective, as average healthcare expenses for people with overweight and obesity are significantly higher than those observed in the rest of the population, resulting in additional annual healthcare costs estimated by Gorasso et al. (2022) at 0.6 percent of GDP. In addition, comorbidity related to obesity and overweight results in increased absenteeism that costs the Belgian economy an estimated 0.2 percent of GDP.

• 2)

  Appropriate use of resources

Numerous studies suggest that health resources are not optimally used in Belgium. For example, though it is more developed than the screening of other forms of cancers (OECD Observatory on Health Systems and Policies, 2021), breast cancer screening is not yet sufficiently widespread to be cost-effective, and age groups that are not part of the target group are overrepresented in the population screened (Devos et al., 2019; INAMI-RIZIV, 2022). Reducing unjustified variations in medical practices (documented on the government website Healthybelgium.be) could also reduce healthcare costs without affecting health outcomes. In some cases, it could even improve them. Some studies argue that Belgium should reduce the overuse of antibiotics and psychotropics, and correct the inadequate prescription of antidepressants and anticholinergic drugs (Devos et al., 2019; Gerkens and Merkur, 2020; HealthyBelgium, 2022; INAMI-RIZIV, 2022). Also, medical imaging appears to be used more in Belgium than in peer countries (Figure 23) and in some cases (e.g., scans of lumbar spine) is medically unnecessary (INAMI-RIZIV, 2022).

• 3)

  Pharmaceuticals

Promoting the use of generics could result in savings. Despite policies put in place (e.g., incentives for physicians to prescribe a certain percentage/quota of low-cost medicine and the convention to stimulate the use of biosimilars - Gerkens and Merkur, 2020), the use of generics remains low, with a share both in volume and in value, below that of other EU countries (Figure 24). The share of generics is particularly low in hospitals, where it accounts for 3.5 percent in value and 17.5 percent in volume compared to 17.5 percent and 44.2 percent respectively, in six other EU countries, for which data are available. In this context, the conclusions of an official working group that recently reviewed the “purchasing policy” for medicines in hospitals may lead to measures to ensure that generics are part of the hospital tenders.

Increasing competition could also reduce the cost of pharmaceuticals. Community and hospital pharmacies have a monopoly on the distribution of pharmaceuticals (prescription and over-the-counter). As a result, pharmaceuticals are not available via supermarkets, and Internet sale is only possible for registered over-the-counter pharmaceuticals if a pharmacy owns the website (Gerkens and Merkur, 2020). Reducing the cost of drugs through greater competition would improve access to healthcare, as pharmaceuticals are an important share of out-of-pocket spending (OECD Observatory on Health Systems and Policies, 2021).

A review of the price-setting mechanism and greater transparency are warranted. In some cases, the price of drugs can be negotiated bilaterally between the authorities and the pharmaceutical producer managed entry agreements or MAEs. Most of the time, MAEs include financial compensation mechanisms that are confidential. The share of pharmaceuticals under MAEs increased from 13 percent of expenditure in 2014 to 31 percent in 2018, leading Gerkens and Merkur (2020) to conclude that they “have become the rule rather than the exception” and that “the exponential increase in the price of innovative treatments and the lack of transparency in confidential price agreements threaten the system.”

• 4)

  Reliance on hospitals

Hospital services are an important reason why Belgium spends more than peers on healthcare. In 2019, the difference in public health spending between Belgium and the EU27 was entirely due to the difference in spending on hospital services. More than half of the increase in public healthcare spending during the pandemic was related to hospital services, compared to slightly over 35 percent in the EU. As a result, in 2021, 49 percent of public health spending was on hospital services; 7 ppts more than in the EU27 (Table 6).

High usage of hospital services is partly related to financial incentives and hospital organization. The reimbursement system provides patients with incentives to use hospital services. This contributes to a relatively high rate of avoidable hospitalization. Avoidable hospitalization for some diseases, such as asthma or chronic obstructive pulmonary disease, is higher than the EU average, highlighting that there is room to improve primary care to manage chronic diseases (OECD Observatory on Health Systems and Policies, 2021). It also points to the importance of developing integrated care and removing financial obstacles to day-hospitalization for hospitals and patients (such as complex payment rules and lack of transparency). These policies would also help continue reducing the average length of stay (Figure 25) and health costs, since a shorter stay reduces the cost per discharge and shifts care from inpatient to less expensive post-acute settings, all other things being equal.

Policies to increase efficiency would free resources that would help absorb the cost of aging and should be, at least partly, used to prepare the health system for the medical challenges related to aging. Because healthcare needs of elderly patients differ from the needs of younger patients, aging has large implications that require reforms of the healthcare system. The main adjustments concern hospital organization, human resources, and long-term care needs.

• 1)

  Preparing hospitals for the challenges of aging

Aging will increase demand for hospital services. Elderly patients’ use of hospital services is already high. For example, patients 85 and older already represented 9.2 percent of hospital patients in 2011 and 16.25 percent of hospitalizations (Deschodt et al., 2015). Moreover, elderly patients are usually associated with a higher risk of admission through the emergency department and with a longer hospital stay (De Foor et al., 2020; Deschodt et al., 2015; Van de Voorde et al., 2017; Vilpert et al., 2013).

The expected increase in demand for hospital services requires a reorganization of hospitals to mitigate its cost. Belgium has a higher density of hospital beds than most EU countries (Figure 26), but they need to be reoriented toward geriatric units. Overall, the healthcare system has a longstanding over-capacity in most types of beds but an under-capacity in geriatrics (Deschodt et al., 2015; Van de Voorde et al., 2017). As a result, in 2011, more than a quarter of hospitalized patients older than 75 were in non-geriatric acute care units, and they accounted for 43 percent of all hospitalization days in acute non-geriatric units (Deschodt et al., 2015). In the short term, given the decline in the projected need for traditional beds, reconverting some of the pediatric, and maternity, surgical, or internal medicine beds into geriatric beds would reduce the fiscal cost of increasing capacity for geriatric care (Van de Voorde et al., 2017; Gerkens and Merkur, 2020). Despite efforts to convert existing facilities and promote home care, the rapidly growing demand for geriatric beds will likely exceed these solutions in the long term. As a result, additional healthcare spending appears inevitable (HealthyBelgium, 2022).

The expected increase in demand for hospital services also requires a reorganization of medical processes and increased cooperation, which may have a fiscal cost. Van de Voorde et al. (2017) acknowledge that many elderly patients are not admitted to geriatric units due to capacity constraints. However, they also note that elderly patients often require treatment for multiple medical conditions, some of which are appropriately handled in non-geriatric units. As elderly patients suffer more from comorbidities than younger patients, their treatment is complex and implies — in addition to a greater utilization and reorganization of hospital resources and medical processes — a need for efficient coordination. This coordination goes beyond hospitalization, as a patient’s stay in the hospital can be prolonged if the discharge is delayed due to a lack of coordination with rehabilitation institutions or nursing homes (De Foor et al., 2020). This also highlights that, to reduce the aging pressure on hospitals, investment in long-term care facilities is needed.

Increasing spending on prevention could help mitigate the cost of aging if it allows the elderly to maintain healthier lives for longer. Belgium has room for improvement, as the share of the elderly population in need of care is among the highest in Europe (Figure 27). In this context, it is important to note that the health status of an elderly person at a certain age may be better in several decades than it is now, which could lower the cost of aging (Kotschy and Bloom, 2023).

• 2)

  Ensuring an adequate geriatric-support workforce

The health needs of an aging population will require different skills from physicians and nurses. Belgium already faces a shortage of physicians and nurses with expertise in elderly care. This shortage will grow as the number of elderly patients increases and aging intensity rises (Figure 3). Therefore, although they imply a cost in the short-term, training policies need to be implemented rapidly so that the increase in geriatric demand is met and the increased hospital capacity is staffed by a sufficient and qualified geriatric workforce (Van de Voorde et al., 2017).

• 3)

  Ensuring adequate long-term care

Despite more prevalent informal long-term care than in other OECD countries (OECD, 2017; Devos et al., 2019) and a long-standing policy to develop at-home care services (HealthyBelgium, 2022), which both tend to postpone institutionalization, public spending on long-term care is among the highest in Europe (Figure 28). The OECD (2023) projects that spending on long-term care will increase by about 2 percent of GDP by 2070. Such an increase again poses fiscal and operational challenges to increase long-term capacities, both in terms of beds and in terms of adequate staffing.

Belgium’s rapid population aging constitutes an important fiscal challenge. As all available projections point to a substantial increase in the cost of aging in the current decade, the sustained fiscal consolidation needed to reduce the debt level and rebuild fiscal buffers post-COVID will face significant headwinds.

Belgium’s pension spending is increasing significantly and rapidly, requiring policy measures. Notably, EC and national projections concur that pension spending during the current decade will increase by 1.75 percent of GDP. Therefore, additional spending containment measures (to increase labor force participation, notably of older workers; to tighten eligibility conditions; and to avoid an increase in the generosity of pensions) are needed now to allow a meaningful fiscal consolidation over the coming years. To ensure the sustainability of the pension system and its important role in the Belgian social model, an increase in social contributions could also be considered, though it would run against the policy of reducing labor taxation.

Access to healthcare and health outcomes are comparatively good in Belgium. However, they are achieved at a relatively high cost, suggesting large potential efficiency gains. Realizing these efficiency gains will require significant reforms — more emphasis on prevention, reforming the organization and role of hospitals, promoting the use of generic drugs, and ensuring a more appropriate use of costly healthcare resources — that would free resources. These resources could help absorb part of the projected increase in fiscal spending on health due to aging but also to help prepare the health system for the medical challenges related to aging patients. Indeed, because healthcare needs of elderly patients differ from the needs of younger patients, aging has large implications that require costly reforms. The main reforms concern hospital organization and human resources, as well as preparing for the increased demand for long-term care.