Journal Issue

Bulgaria: Selected Issues Paper

International Monetary Fund. European Dept.
Published Date:
November 2016
  • ShareShare
Show Summary Details

Fiscal Implications of Demographic Changes in Bulgaria1

A. Introduction

1. Bulgaria’s population has been shrinking since 1984 due mainly to emigration and declining fertility rates, leading to its high old-age dependency ratio2 (Figure 1). Bulgaria’s population trend is in contrast to population developments in Europe—including advanced and developing Europe—where population has continued to grow over the last 50 years.

Figure 1.Europe: Population (In million)

Source: UN 2015

2. While demographic projections are surrounded by significant uncertainty, the old-age dependency ratio is expected to increase in the coming decades in all the European countries (Figure 2). At around 30 percent, Bulgaria’s old-age dependency ratio is currently the highest among Central, Eastern, and Southeastern Europe (CESEE), and comparable to that in the Eurozone. According to the United Nation’s medium-fertility scenario, Bulgaria’s old-age dependency ratio is expected to peak around 2055 at about 54 percent, which would then decline to below 50 percent by 2100. Such a projection is in contrast to the projection in Albania and Turkey where the old-age dependency ratio is currently low (11 percent for Turkey and 18 percent for Albania) and but would increase rapidly in between 2055 and 2100.

Figure 2.Change in Old-Dependency Ratio

(In percent)

3. Demographic changes can pose significant fiscal challenges going forward. Population aging, stemming from declining fertility and increasing longevity and emigration, is expected to raise the old-age dependency ratios and shrink the working age population. The accompanying surge in age-related spending, lower economic growth, and potentially lower government revenues, could place considerable pressures on public finances. While the near-term fiscal effects of these demographic developments are likely to be small, their long-term fiscal implications could be significant. Moreover, given the uncertainty surrounding demographic projections, an even faster transition to a declining workforce cannot be ruled out. Therefore, policy-makers in countries affected by adverse demographic dynamics need to be cognizant of fiscal risks from demographic developments

4. A shrinking population could have a large direct impact on budgetary expenditures and revenues.

  • Expenditures. Spending on old-age pensions and health and long-term care would rise. Because pensions target specific age groups, pension spending increases as the share of the population within these age groups rises. Although healthcare is universal across age groups, per capita health spending tends to increase with age. Therefore, aging will increase public health spending as well. Unlike pensions and healthcare, the potential impact of demographic changes on other types of spending (e.g., education) tends to be uncertain.
  • Revenues. Individual income and consumption spending patterns tend to change over the lifecycle. For example, workers’ incomes tend to increase from early in their careers until the middle age and decline towards retirement. Consumption spending follows a similar pattern, although in the very early years of one’s career consumption may exceed income (Thurow, 1969). Government revenues are likely to reflect these developments as well as changes in the overall size of the population. More generally, lower economic growth, following a slowdown in population growth and a shift in the age-gender structure towards less active cohorts, would translate into lower revenues, assuming unchanged policies.

5. The aim of this paper is to simulate the fiscal impact of demographic changes in Bulgaria using a simple framework recently published by the Amaglobeli and Shi (2016). The framework allows for a granular approach in projecting the impact of demographic changes—both the composition and size of population—on output by taking into account differences in labor force participation and employment rates by age group and gender. Regarding demographic projections, this paper relies largely on those from the United Nations (2015 Revision of the UN World Population Prospect). The UN 2015 reflects past demographic trends and country-specific information to inform projections, taking into account all currently available information. Nonetheless, these projections must be taken with caution. Like past projections have been subject to large errors, future realizations of fertility, mortality, and migration rates may differ substantially from the predicted levels. Given the significant uncertainties associated with population projections and other factors driving growth, the main aim of the paper is not to accurately project additional fiscal burden from the aging population, but rather to simulate how sensitive the results could be depending on demographic assumptions under the assumption of no policy reaction.

6. There are two main contributions in this paper, over and above the analysis in the European Commission’s Aging Report. First, using the UN population projections allows for a longer projection period—through 2100—compared to the EC’s Aging Report, which covers projections through 2060. Good public finance practices suggest fiscal projections, including fiscal risks, for a period that covers an individual’s life time. Given the current life expectancy of around 80 years old, 45-year projections covered by the EC’s Aging Report may not be sufficiently long. At the same time, extending a projection horizon increases uncertainties, therefore results—especially those closer to the end of the projection horizon—should be interpreted with caution. Second, this paper factors in a relatively per capital public healthcare spending of the elderly to that of the prime age groups in the baseline scenario based on the experience in OECD countries (Figure 3). The EC’s Ageing Report assumes such excess cost as one of the alternative scenario.

Figure 3.Europe: Public Healthcare Expenditure

(In percent of GDP per capita, sample median)

Note: Sample medians of 20 OECD countries.

Source: De la Maisonneuvet al. (2013)

7. The structure of the paper is as follows: Section B reviews Bulgaria’s pension system, including the 2015 reforms. Section C discusses in detail the analytical framework to estimate fiscal implications of aging population, including the estimation of potential output under different demographic scenarios and long-term fiscal spending on pension and healthcare. Section D discusses key takeaways and recommendations.

B. Bulgaria’s Pension and Healthcare System

8. Bulgaria currently has a three-pillar contributory pension system supplemented with noncontributory social pension scheme (Figure 3). Earners—i.e., the employed, self-employed, and farmers—born after 1959 are obliged to contribute 17.8 percent of their social security income to Pillar 1 defined benefit public pension system and Pillar 2 defined contribution private pension system. As discussed later (Paragraph 6), from the beginning of 2016, earners can choose whether to contribute to both Pillar 1 and Pillar 2 or to Pillar 1 only (Paragraph 10).

Bulgaria: Multi-Pillar Pensions System in 2016

9. Bulgaria’s current public pension system is characterized as having relatively low (i) low statutory retirement ages; (ii) pension contributions; and (iii) benefit ratio. At 63.7 years for men and 60.7 years for women, Bulgaria’s current statutory retirement ages are below the EU medians of 65 years for men and 63.5 years for women. The collection of contributions to the public pension were 7 percent of GDP in 2013, below the EU median of about 8 percent of GDP, and Bulgaria’s benefit ratio—average pensions in relation to average wages—is also low. The low contributions and benefit ratio are due mainly to the existence of Pillar 2 to which serves as mandatory savings. Indeed, the benefit ratio is comparable to the countries that also have Pillar 2 schemes while the contributions are on a high end among them.

Figure 4.Public Pension Contribution and Benefit Ratio

Note: Those in yellow, including Bulgaria, have a mandatory private pension scheme.

Source: 2015 EU Ageing Report

10. To address the sustainability concerns of the pension system from aging population and migration while enhancing the adequacy of pension benefits, Bulgaria implemented pension reforms in 2015. The reforms consist of (i) parametric reforms to the Pillar 1 public pension system and (ii) the introduction of an option for unlimited shifts between the Pillar 1 and Pillar 2 pension systems.

  • Parametric reforms to the Pillar 1 pension system. These include (i) increasing the social security contribution rate by 2 percent by 2018, (ii) raising the retirement age to 65 years by 2029 for men and by 2037 for women, and (iii) extending the minimum required service length to 40 years for men and 37 years for women by 2027; and (iv) increasing the accrual rate—the weight of one year of service—gradually from 1.2 to 1.5 by 2026.
  • Introduction of an option for unlimited shifts between Pillar 1 and 2 pension systems. The option allows an individual to unlimitedly shift between Pillar 1 and 2 pension systems through five years before the retirement. Regardless of the time vested in Pillar 2, an individual whose final decision is to permanently shift to the Pillar 1 system will receive undiscounted amount of pension benefits from Pillar 1. In such case, the pension benefit would be broadly equivalent to 60 percent of his social security income.

11. The 2015 pension reforms are expected to moderately reduce public pension deficits in the next decade, but do not address long-term sustainability concerns. Although the parametric reforms are expected to reduce Pillar 1 annual deficits moderately in the next decade, the expected increase in the income replacement rate of pension benefits will likely more than offset the combined positive impact from raising the contribution rate and retirement ages afterwards (Figure 5). Even after the reforms, the National Social Security Institute (NSSI)—which manages the state social security including the Pillar 1 pension—is expected to carry deficit, indicating a continued need for additional budget transfers every year.3 The newly-established option guarantees full Pillar 1 benefits, suggesting a shift of downside risks from individuals to the public system.

Figure 5.Bulgaria: Financial Flows of Public Pension 1/

(Percent of GDP)

1/ Assuming no shift to Pillar 1 from Pillar 2,

source: NSSI

12. In addition, relatively inefficient healthcare system in Bulgaria could augment fiscal challenges as the demand increases in line with population aging. Bulgaria’s healthcare system is generally less efficient compared to the frontier mainly because of underutilization of preventive measures, excess use of costly inpatient services, and inefficient pharmaceutical pricing. Fragmented legislations on long-term care services also create inefficiency in the system.

C. Long-Term Fiscal Implications of Demographic Changes

Analytical Framewor

13. The framework has two components: estimating the growth impact of demographic changes, and projecting age-related spending. The basic unit of the analysis is a five-year cohort from the disaggregation of the population by age and gender. The focus on five-year age-gender cohorts allows to capture the life-cycle behavior of labor supply: low for youth, increasing and flattening during prime age, and decreasing closer to retirement. This life cycle pattern can also differ for men and women. Projected changes in long-term output and available long-term population projections are then used to estimate government revenues and expenditures and derive net fiscal effect.

14. Several demographic scenarios—both upside and downside—are considered.

  • The UN scenarios. There are eight different scenarios published by the UN based on different assumptions on fertility, mortality, and international migration: the scenarios include medium, high, low, and constant fertility, instant-replacement-fertility, zero-migration, constant-mortality, and no-change (i.e., constant-fertility and constant-mortality).4 The UN’s central projection is the medium-fertility variant scenario, which can be considered as the most probable scenario. The scenario assumes a decline of fertility for countries where large families are still prevalent while a slight increase of fertility in countries with fewer than two children per women on average. In addition to the medium-fertility scenario, this paper considers low- and high-fertility scenarios from the UN database.
  • Bulgaria-specific scenario. Given Bulgaria’s unique challenge from brain drain, emigration of work-age population: more concretely, population aged 15-64 is assumed to be 1 percent below the medium-fertility scenario throughout the projection period. The emigration of the work-age population is also assumed affect the total productivity.

Estimating the GRowth Impact

15. A production function approach is used to estimate long-term output, taking into account demographic developments. A key intended contribution of this paper is to endogenize the effect of demographic change on output growth. Following the standard specification of the Cobb-Douglas production function with constant returns to scale, output can be expressed as the product of basic factor inputs and productivity. Specifically,

Where, Yt is real output, TFPt is the total factor productivity, Kt is the stock of capital, Lt is the aggregate labor, and β is the share of labor in output. Following EC (2012) and IMF WEO (2015), age-gender cohort specific information is used to decompose aggregate employment (Lt) into four elements. In particular, rewriting equation (1) in logarithmic terms, output is expressed as:

where j indicates the age-gender cohort, N is the number of individuals in each cohort, LFP and E denote cohort-specific labor force participation and employment rates, respectively, and w is the weight factor to adjust for the difference between number of employees and the effective units of labor supplied.5

16. Equation (2) can be used to obtain historical TFP estimates.6 Historical labor force participation and employment rates for each age-gender cohort are combined with the evolution of the size of each individual cohort to estimate aggregate labor. The coefficient β can be obtained by calculating the share of wage bill in GDP.7 Capital stock is estimated using the perpetual inventory method, with the capital stock in each period equal to net capital formation plus the estimated stock in the previous period (equation 3).

17. Using equation (2) and TFP estimates, long-term projections for potential GDP are calculated under different demographic scenarios. Aggregate labor is assumed to evolve in line with total population and its age-gender composition under different demographic scenarios. The age- and gender-specific labor-force participation rates up to age 54 are assumed to remain unchanged from the 15-year historical averages while the participation rates for age 55 and above are assumed to rise as the statutory retirement ages increase.8 Yet, the labor force participation rates for older age cohorts tend to remain below those for younger age cohorts, therefore population aging will contribute to the reduction in aggregate labor force participation. Abstracting from policy changes, TFP is assumed to grow at its three-year historical average, while capital accumulation is assumed to follow the balanced-growth condition:

The share of labor in output (β) is assumed to gradually increase from the three-year historical average (0.35) to 0.37 by 2100.9

Estimating Age-Related Spending

18. Pension spending as percent of GDP is decomposed into four key elements and expressed as an identity (following Clements and others, 2014). These components are: (i) the replacement rate (RR), which is calculated as the average pension over average output per worker; (ii) the coverage ratio (CR), which measures the share of pensioners in the total population above the retirement age (above 65); (iii) the old age dependency ratio (ODR), which is measured as the ratio of population above 65 and the working age population (15-64); and (iv) the inverse of labor participation (LP), defined here as the share of workers in the total working age population—different from the definition of labor force participation in (2).


Each of the four terms in (5) is influenced either by demographics or by policy changes. This paper uses implied replacement rate (RR) and coverage ratio (CR) calculated with the authorities’ baseline projections on pension spending in percent of GDP and the UN mid-fertility demographic projections.10 The remaining variables are determined by demographic assumptions.

19. Healthcare spending is expressed as the product of three elements (following Clements and others, 2015). These are: (i) relative generosity of the healthcare package for the younger population expressed as per capita health cost for younger population divided by the per worker GDP (GHP); (ii) the inverse of the labor participation (LP) rate for population between 0 and 64 years of age; and (iii) the dependency ratio (DR) times a coefficient (𝛼). The latter measures the ratio of per capita health cost for older population over per capita health cost for younger population.


Similar to pension expenditure projections, each term of (6) is affected either by policy changes or by demographics. In addition to its direct effect through changes in dependency ratio and labor participation rates, population aging also affects the generosity of the healthcare package indirectly through growth. The coefficient α is exogenously determined and could be calculated from available sources. Data on public health care expenditure is available for a group of OECD countries from de la Maisonneuve and others (2013). If no country specific information is available, the average for countries at similar income levels could be used.11 To isolate the demographic effects, the generosity of healthcare package (historical data) and the α coefficient are kept constant across scenarios, implying no policy measures.

Assumptions Underpinning Estimates

  • The capital stock depreciation rate, which is used for the historical growth accounting exercise, is set at 5.9 percent per year in line with Nadiri and Prucha (1993).
  • Long-term non-age expenditure elasticity to nominal GDP is set at one.
  • The default ratio of per capita health spending of individuals above older than 65 years relative to those younger than 65 years is 3.5, which is the average for a group of OECD countries for which data was available (De la Maisonneuve and others, 2013). Alternatively, estimates of country specific ratios from the country authorities could be used.

Put them together

20. Bulgaria’s growth prospect depends largely on the assumption of future productivity and demography (Figure 6).12

  • Baseline (mid-fertility) scenario (left panel). The scenario assumes the UN’s medium fertility rate, the three-year historical average of TFP growth rates of 1.1 percent, the balanced growth of capital, and a gradual increase in the share of labor in output. Under this scenario, Bulgaria’s potential growth rate would decline from 2½ percent in 2020 to 1⅔ percent by 2055 then return to around 2½ percent by 2100.
  • Low-fertility scenario (middle panel). The scenario assumes the UN’s low-fertility rate, the three-year historical average of TFP growth rates of 1.1 percent, the balanced growth of capital, and a gradual increase in the share of labor in output. Under this scenario, Bulgaria’s potential growth rate would decline from 2½ percent in 2020 to ½ percent by 2080 then return to around 1¼ percent by 2100.
  • Mid-fertility with low TFP growth scenario (left panel). To show the sensitivity of growth projections to the assumption of TFP growth rates, the scenario assumes the UN’s medium-fertility rate, zero TFP growth, the balanced growth of capital, and a gradual increase in the share of labor in output. Under this scenario, Bulgaria’s potential growth rate would remain negative (between around -½ and -1½ percent) through 2100.

Figure 6.Long-Term Real GDP Growth Projections

1/ Assuming TFP to grow at a quarter of the three-year average growth rates.

21. The fiscal implications vary greatly depending on demographic projections, as well as the nominal rigidity of pension benefits (Figure 7). All the below scenarios, except the last one, assume the baseline growth parameters defined in Paragraph 17, combined with various demographic assumptions.

  • Baseline scenario. The scenario assumes the baseline growth parameters and the UN’s medium-fertility scenario, which projects a decrease in population to 3.4 million and an increase in the old-age dependency ratio to 49 percent by 2100. Under this scenario, additional annual age-related spending would be 3 percent of GDP by 2055 and 4 percent of GDP by 2100. The increase is mostly driven by health spending, while there is minimal increase in pension spending projected thanks to the recent parametric reforms.
  • Low-fertility scenario. The scenario assumes the baseline growth parameters and the UN’s low-fertility scenario, which assumes a decrease in population to 1.8 million and an increase in the old-age dependency ratio to 78 percent by 2100. Under this scenario, additional annual age-related spending would increase by 4 percent of GDP by 2055 and 11 percent of GDP by 2100.
  • High-fertility scenario. The scenario assumes the baseline growth parameters and the UN’s high fertility scenario, which assumes a decrease in population to 5.6 million and an increase in the old-age dependency ratio to 35 percent by 2100. Under this scenario, additional annual age-related spending would increase by 2 percent of GDP by 2055, which would then decline to 1 percent of GDP by 2100.
  • High-migration scenario. The scenario assumes the baseline growth parameters, population aged 15-65 to be below the UN’s mid-fertility scenario by 1 percent, and TFP growth rate to be below the baseline by 5 percent. Under this scenario, additional annual age-related spending would be generally same as the baseline scenario.

Figure 7.Macro-Fiscal Variables Under Various Population Scenarios

22. The impact of age-related spending on government debt depends on long-term growth assumptions.13 By construction, two scenarios with same demographic assumptions yield same additional age-related spending in percent of GDP even if their growth projections differ owing to different productivity assumptions. For example, the baseline scenario (medium fertility) and a scenario with the medium-fertility low TFP growth rates project same additional age-related spending in percent of GDP. However, those two scenarios have substantially different debt implications due to their different debt dynamics driven by different GDP growth rates (Figure 8). For all the scenarios, government debt is projected to gradually decline during the next 10-20 years, which will start rising thereafter. Under the baseline scenario, government debt would decline from 30 percent of GDP in 2016 to around 16 percent of GDP by 2035, which will then increase to above 60 percent of GDP by 2064 and close to 100 percent of GDP by 2100. Under the low-fertility scenario, debt would exceed 60 percent of GDP by 2059, followed by a rapid increase to above 250 percent of GDP by 2100. Under the high-fertility scenario, debt would peak at around 45 percent of GDP in the 2060s, which would then start declining toward 30 percent of GDP by 2100.

Figure 8.Bulgaria: Government Debt under Different Growth and Demographic Scenarios

(Percent of GDP)

23. The newly-established option for unlimited shifts between Pillar 1 and Pillar 2 could further increase fiscal pressures. According to the authorities’ estimates, if 100 (50) percent of the workers decide to shift to Pillar 1 in 2016, Pillar 1 pension spending would increase by 3 (1.5) percent of GDP compared to Pillar 1 pending without any shifts by 2060. However, the net fiscal impact of the shift scenarios is positive compared to the no-shift scenario in the medium term because the government will receive additional social security contributions from the individuals who shift to Pillar 1. Assuming the baseline growth projections, government debt in percent of GDP for both shift scenarios is projected to remain below that for the no-shift scenario through 2050, but exceed thereafter. Government debt could be higher than the no-shift scenario by around 12 (6) percent of GDP by 2060 and 33 (16½) percent by 2100 assuming 100 (50) percent shift to Pillar 1 in 2016 (Figure 9).

Figure 9.Bulgaria: Government Debt under Different Pillar 1-2 Compositions

(Baseline Growth Projections, percent of GDP)

1/ Underlying growth projections are based on the UN’s medium-fertility scenario with the assumptions of the historical labor-capital ratio and improving TFP (at the two-thirds of the historical peak).

Sources: NSSI, IMF staff estimates

D. Key Takeaways and Policy Recommendations

24. Given the significant uncertainties with demographic projections and the risks related to additional aging related fiscal spending, Bulgaria would benefit from further reforms to the pension system. Several options can be considered:

  • Raise the social security contribution rate. To prevent an increase in old-age poverty, the social contribution rates could be raised further while paying attention to potential impact on competitiveness.
  • Increase the statutory retirement ages. Bulgaria’s statutory retirement ages for men and women are expected to remain below the EU medians even after reaching the statutory retirement age of 65 years for both men and women by 2037. To leave scope for further increases in retirement ages, the 2015 pension reforms introduced the concept of automatic link between the statutory retirement ages and life expectancy once the statutory retirement ages reach 65 years. However, to de-politicize further parametric reforms, the modalities of adjusting the retirement ages (e.g., what conditions trigger a revision of the statutory ages, what coefficients determine new retirement ages) need to be fleshed out. Raising the retirement age, however, should be accompanied by various measures—such as active labor market policies or wage subsidies—to ensure old-age employment. In addition, the government should be alert that increasing the retirement age could disproportionately affect low-income earners’ beneficiary period given their typically shorter life expectancy, thereby reducing the progressivity of public pension systems.
  • Improve the viability of Pillar 2 and 3 private pension schemes. In response to a general concern about the viability of the Pillar 2 scheme, the government (FSC) has initiated a review of Pillar 2 assets, which is expected to help the government identify the weaknesses of the system and policy options going forward. Sound defined-contribution pension schemes will help prevent old-age poverty, or large adjustment in lifestyle or consumption after retirement while minimizing the sustainability concerns of public finances.

25. Further pension reforms, however, should strike a right balance between ensuring fiscal sustainability and preventing old-age poverty. In particular, raising the retirement age should be accompanied by various measures—such as active labor market policies or wage subsidies—to ensure old-age employment. In addition, the government should be alert that increasing the retirement age could disproportionately affect low-income earners’ beneficiary period given their typically shorter life expectancy, thereby reducing the progressivity of public pension systems.

26. Boosting long-term growth through enhancing fertility rates, promoting labor force participation, and mitigating brain drain can help mitigate fiscal pressures from aging population. Bulgaria’s labor-force participation rate is below the EU median although its participation gap between male and female (including that for fulltime) is among the lowest. The gradual increase in the statutory retirement ages would increase the participation rate. At the same time, polices to encourage people to participate in the labor market (e.g., provision vocational training) should also be explored. Promoting child-care support would help prevent a decline in the fertility rate from an increase in labor-force participation. Return migration can yield significant benefits by bringing back skills and contributing to the diffusion of organizational and technical knowledge acquired by emigrants abroad. Policies should focus on removing barriers to reintegration of return migrants into the workforce, including, by recognizing foreign credentials and experience, and opening access to the service sector by deregulating professions.

27. The newly-established option for unlimited shifts between Pillar 1 and 2 systems—that was introduced due to public concerns about the performance of the Pillar 2 system—raises a number of concerns.

  • The reform will increase budget uncertainty. For example, unpredicted shifts from Pillar 1 to Pillar 2, due to concerns about the viability of private pension funds, would require unbudgeted state transfers to the NSSI, adding complication to the public finance management. Uncertainty also remains in the management of the Pillar 2 assets transferred to the Silver Fund. If mismanaged—for example, used for financing budgetary expansion in the short run—the sustainability of Pillar 1 would be jeopardized.14
  • Abrupt large shifts are anticipated when the economy is in an extreme situation, such as a sharp economic downturn or an asset pricing boom. People will decide on whether to switch to Pillar 1 or not by comparing the amount of Pillar 2 assets to give up and projected additional Pillar 1 benefits to receive from switching (see Annex 1). This would suggest that in normal times, frequent switches between Pillar 1 and 2 may be limited especially given no interest will be earned on the assets transferred from Pillar 2 to the Silver Fund (people have less incentive to shift back to Pillar 2 after leaving their assets with zero earnings).15 When the economy experiences a sharp recession where the performance of financial markets plummets, people could shift from Pillar 2 to Pillar 1 as investment performance drops and preference to rely on the public sector increases. This would suggest that the government is absorbing the downside risks and standing behind Pillar 2.
  • The reform could thwart the productivity of defined-contribution schemes during a period when private pension funds are already facing substantial challenges given the unfavorable demographic changes projected in the coming decades.
    • a) Unlimited switches could create a fiscally costly quasi-competition between the state and the financial sector: the scheme creates perverse incentives for the government to try to attract more contributors to the public system to gain in the short run, but at a cost in the long run, generating long-term cost potentially exceeding short-term fiscal gains. There may also be incentives for the government to regulate private pension schemes in a way to disadvantage them.
    • b) Uncertainty associated with the size and timing of switches could make Pillar 2 asset management more complicated, and would also make long-term sustainability analyses of Pillar 1 less reliable.

28. Continued efforts are also need to contain healthcare costs, especially related to pharmaceutical and long-term care. Measures to address Bulgaria’s low use of preventive measures and outpatient services and overuse of inpatient care could improve the health outcomes. In addition, recently-introduced measures to address the over-supply of hospitals treating a low number of patients and contain pharmaceutical pricing would help. Demand for long-term care services is bound to increase strongly with aging. Providing high-quality long-term care services while ensuring financial sustainability would call for a legislative amendment to enhance a synergy between social services system and health care system, including private provision of long-term care services. Introducing a long-term care insurance program is also an option.


    AmaglobeliDavid and WeiShi. 2016. “How to Assess Fiscal Implications of Demographic Shifts: A granular ApproachHow-to notes 2Fiscal Affairs DepartmentWashington, DC.

    • Search Google Scholar
    • Export Citation

    ClementsBenedictKamilDybczakVitorGasparSanjeevGupta and MauricioSoto. 2015. “The Fiscal Consequence of Shrinking Populations.” IMF Staff Discussion Note SDN/15/21International Monetary FundWashington, DC.

    • Crossref
    • Search Google Scholar
    • Export Citation

    De la MaisonneuveChristine and Joaquim.Oliveira Martins. 2013. “Public Spending on Health and Long-Term Care: A New Set of Projections.” OECD Economic Policy Paper 06Organisation for Economic Co-operation and DevelopmentParis.

    • Crossref
    • Search Google Scholar
    • Export Citation

    European Commission. 2012. Fiscal Sustainability Report 2012. European Economy 8|2012. Brussels: Economic and Financial Affairs, European Commission.

    • Search Google Scholar
    • Export Citation

    European Commission. 2015. The 2015Ageing Report—Underlying Assumptions and Projection Methodologies. Brussels: Economic and Financial Affairs, European Commission.

    • Search Google Scholar
    • Export Citation

    FeyrerJames. 2007. “Demographics and Productivity.” Review of Economics and Statistics89(1): 1009.

    GöbelChristian and ThomasZwick. 2009. “Age and Productivity—Evidence from Linked Employer Employee Data.” ZEW Discussion Paper 09-020Center for European Economic ResearchMannheim, Germany.

    • Search Google Scholar
    • Export Citation

    HigginsMatthew. 1998. “Demography, National Savings, and International Capital Flows.” International Economic Review39(2): 34369.

    International Monetary Fund. 2015. “Uneven Growth: Short- and Long-Term Factors.” Chapter 3 in World Economic Outlook. WashingtonDCApril.

    • Search Google Scholar
    • Export Citation

    NadiriIshaq and IngmarPrucha. 1993. “Estimation of the Depreciation Rate of Physical Capital and R&D Capital in the U.S. Total Manufacturing Sector.” NBER Working Paper 4591National Bureau of Economic ResearchCambridge, MA.

    • Search Google Scholar
    • Export Citation

    ThurowLester. 1969. “The Optimum Lifetime Distribution of Consumption Expenditures.” American Economic Review59(3): 32430.

1Prepared by Aiko Mineshima.
2Defined as the ratio of population ages 65 and older to population ages 15 to 64.
3Budget transfers to the NSSI have been twofold: (i) ex-ante annual regular transfers of 12 percent of projected insured income, and (ii) ex-post additional transfers to finance the NSSI’s deficit. Starting from 2016, there will be a single item called “transfer for shortage,” which combines the above two types of transfers.
4The eight scenarios include medium, high, low, and constant fertility, instant-replacement-fertility, zero-migration, constant-mortality, and no-change (i.e., constant-fertility and constant-mortality).
5Average hours worked for each cohort can be a proxy for the weight factor. Other desirable weights would be to adjust for labor productivity, possibly utilizing the average education attainment for each cohort as a proxy. The literature shows that labor productivity significantly differs across age groups (e.g., Göbel and Zwick, 2009, Skirbekk, 2004). Due to the lack of data feasibility for Bulgaria, this paper assumes 1 for the weight factor for all the cohorts.
6It should be noted that TFP is measured as a residual, and any measurement errors in the labor and capital series will be captured in the estimate of TFP.
7Alternatively, this parameter could be obtained from the Penn World Tables or other country-specific estimates.
8More concretely, for example, the labor force participation rate of age 55-59 is assumed to rise to the rate of age 50-54 by 2030 for men and by 2035 for women. The rate of ages 60-64, 65-69, and 70-74 are also assumed to rise to the level of the one notch younger cohort by 2030 for men and 2035 for women and remain constant thereafter.
9These are clearly simplifications, as demographic factors could also affect productivity (see, for example, Feyrer, 2007) and investment (see, for example, Higgins, 1998).
10Since the authorities’ projections are available only through 2060, the implied replacement rate (RR) and coverage ratio (CR) are assumed to stay constant at the implied 2060 levels.
11The 𝛼 coefficient could be estimated as the weighted average of public health care expenditure for age cohorts 65 years and above divided by the weighted average of public health care expenditure for younger cohorts. The relative size of each age cohort could be used as weights. Using this approach, the 𝛼 coefficient is expected to gradually rise over time as the share of older cohorts in the group 65 years and above increases. According to the available country-specific data, public health care per capita GDP costs continuously rise after the age of 45.
12Historical data on labor force participation rates are from Eurostat. Historical and projected data on gross fixed capital formation are from World Economic Outlook. Historical data on compensation of employees are proxies by the nominal average wage multiplied by the number of employed persons from National Statistical Institute.
13Debt projections are calculated with the standard debt equation: Dt = Dt-1 / (1 + nominal GDP growtht) – OBt + SFAt where nominal GDP growth rates are calculated with the assumption of annual inflation of 2 percent while the stock-flow adjustments (SFA) is assumed zero. The overall fiscal balances (OB) for 2022-2100 are assumed to be driven only by the age-related spending, starting from the balanced budget in 2021.
14During periods of economic downturn, the government could be tempted to use the additional assets acquired to the state.
15Transfer of Pillar 2 assets to the Silver Fund should be treated as a financial transaction (below-the-line) in accordance with the Eurostat manual (I.3.4 “Transfer of pension entitlements from the second pillar” of Eurostat’s Manual on Government Deficit and Debt: Implementation of ESA2010).

Other Resources Citing This Publication