Abstract
The paper assesses key aspects of Bulgaria's competitiveness. The challenge is to stay on course and persist with policies that will maintain and strengthen competitiveness. Implementation of the ambitious reform policy with respect to the pension and health care systems is required. The reasons for and implications of low bank credit to the private sector in Bulgaria, and measures to facilitate prudent credit growth are discussed. The statistical data on the economic indices of Bulgaria are also presented in the paper.
II. Pension and Health Reforms in Bulgaria: Restoring Sustainability1
A. Introduction and Summary
By the late 1990s, the pension and health systems in Bulgaria were on a financially unsustainable path and had become highly inefficient. In the case of pensions, adverse demographics, a low effective retirement age, and declining social insurance revenues despite high contribution rates had led to large unfunded liabilities. As for health care, the main problem was a declining quality of services provided at increasing costs to patients. Ambitious reforms of the pension and health systems were initiated in 1999–2000. If fully implemented, these reforms should restore the viability of the social insurance schemes and improve their efficiency.
1. By the late 1990s, the pension and health systems had become unsustainable. The traditional pay-as-you-go (PAYG) pension system administered by the National Social Security Institute (NSSI) was under severe stress: the population was on a declining trend and aging rapidly, and the average effective retirement age had fallen to 56 years owing to the widespread use of early retirement as a social policy tool. Moreover, social insurance revenues had declined substantially during the transition despite high contribution rates, reflecting weak compliance, especially in the emerging private sector (Table A30). As fiscal pressures ruled out large current social security deficits, the authorities had no choice but to allow the real value of pensions to erode: in 1999, the average monthly pension stood at the equivalent of US$35, having declined 60 percent in real terms during the decade. Regarding health, the budgetary costs of the universal health care system were quite low by international standards (Figure 1). This low level was reflected in lack of investment in the sector, contributing to a decline in the quality of services. Moreover, the universal and nominally free system was in practice quite costly to patients as side payments were commonly required. Besides having to deal with the pressing current problems, the authorities were also facing adverse long-term demographic trends which in the absence of major reforms would jeopardize the sustainability of the social insurance schemes (Figure 2).
Bulgaria: International Comparisons
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: National authorities.Bulgaria: International Comparisons
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: National authorities.Bulgaria: International Comparisons
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: National authorities.
Bulgaria: Long-Term Forecasts of the National Social Security Institute
(In thousands)
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI, long-term projection model.Bulgaria: Long-Term Forecasts of the National Social Security Institute
(In thousands)
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI, long-term projection model.Bulgaria: Long-Term Forecasts of the National Social Security Institute
(In thousands)
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI, long-term projection model.2. The pension and health systems had also become nontransparent and overly complex. In the second half of the 1990s, Bulgaria’s social insurance had evolved into a nontransparent system with substantial cross-subsidies and a highly arbitrary and distorted incidence of costs and benefits. Thus, pensions received differed widely among people with equal length of service and lifetime contributions, and periodic ad-hoc adjustments to the level of pensions became the key determinant of their real value. The availability and quality of health care varied with the patient’s employer, location, and ability to provide side-payments.
3. Ambitious pension and health reforms were launched in 1999–2000 to restore sustainability and improve efficiency. The pension reform launched from January 1,2000 aims to restore the long-term viability of the traditional PAYG pension scheme through a significant reduction in entitlements and complementing it with fully-funded components. However, owing to the substantial stock of unfunded pension liabilities, the system will incur losses for some time, as the funded components are built up. The health reform seeks to gradually put in place a mix of private and public providers to deliver health care in a more efficient and equitable manner, albeit at a higher explicit cost to individuals. The new health system is anchored by the Health Insurance Fund (HIF) created in mid-1999 which is envisaged to become an efficient financing and management agency, contracting out health care provision to competing agencies and controlling costs. Health reform will also involve up-front costs to cover institutional capacity building and restoring the long-neglected capital stock.
4. A number of important challenges remain. The blueprint for both reforms is sound, but effective and timely implementation is crucial for the envisaged benefits to materialize. To allow for much-needed additional spending, these reforms should be complemented by improvements in the efficiency of revenue collection. The recent decision by the government to establish a Unified Revenue Agency to collect taxes and social insurance contributions holds the promise of achieving this objective. On pensions, the administrative and regulatory framework for the funded pillars needs to be finalized rapidly, and reducing the transitory costs of reform should be considered by shortening the transition period to the higher retirement ages of 60 for women and 63 for men. As for health care reform, a solution needs to be found for financing hospital care while avoiding a sharp increase in health care costs, and efforts are needed to support the emergence of supplementary health insurance.
5. This study discusses the characteristics of the old pension and health systems, as well as the reform packages. First, it describes the pension system at the end of the 1990s, the reform measures and their fiscal impact, concluding with an overview of key remaining problems. Second, it discusses the health system and its reform in a similar fashion. Third, it provides a comprehensive assessment of the Bulgarian authorities’ policy response to the deep-seated problems in the health and pension sectors.
B. Pension Reform
The old pension system
6. In the 50 years from 1949, pensions in Bulgaria were provided through a traditional defined-benefit PAYG system. The NSSI-administered PAYG system covered all employees in the public and private sector as well as the self-employed. It was financed by payroll contributions and transfers from the republican budget to cover social assistance programs carried out on behalf of the government, and expenditures not covered by the NSSI’s own revenues. The weighted average pension contribution rate, at 42 percent in 1998, was quite high compared with other countries. The retirement age was 60 for men and 55 for women, and workers also had to satisfy specific length-of-service requirements to qualify for old-age pension. Since April 1996, benefits were calculated by multiplying the ratio of the individual’s wage to the average wage, a length-of-service factor, and the average wage in the economy over the preceding three years. However, the formula was temporarily abandoned from 1997 because the resulting benefit levels could not have been afforded. Social pensions of 65 percent of the minimum wage were paid to individuals above age 70 who did not have any other pension. The minimum pension was 90 percent of the social pension, while the maximum was three times the social pension. Survivors’ benefits were available to children and to widows if they were dependent on the worker and over age 50 (60 for men), disabled, or caring for a child under age 16.
7. The parameters of the PAYG system were increasingly out of line with demographic and macroeconomic reality. The retirement age was among the lowest in Europe and proved unsustainable in light of the revenues available. A large number of special privileges added further strains to the system. In particular, special category workers (category I and II labor) could retire early, with their employers paying pension contributions at a higher rate (52 percent for category I and 49 percent for category II workers, compared to 37 percent for category III workers).2 On average in 1998, retiring category I men had clocked up 10 years less working time than their category III counterparts (the average difference for women was 5.8 years). The relative share of special category workers has ballooned over time, and by 1998, categories I and II represented 16 percent of pensioners and 22 percent of pensions paid. The eligibility for early retirement was sharply curtailed at the beginning of 1999, reducing the share of category I and II contributors including the military to around 5 percent of all contributors. On the benefit side, the level of pensions provided were extremely low, and had little relationship with the amount of lifetime contributions, especially following the 1996 change in the benefit formula which was designed to counter the effects of high inflation.
8. Bulgaria experienced adverse demographic developments during the past decade. During the 1990s, a significant wave of emigration resulted in negative population growth, while life expectancy stagnated (Figure 3). The age structure of the population by the late 1990s resembled that in western European countries, with a considerably lower share of young people than in most other transition economies (Figure 4). The ratio of people in pensionable age approached a quarter of the population, and by 1998 pensioners accounted for around 29 percent of the population and 80 percent of those with social security insurance (Table 1).
Bulgaria: Population Dynamics and Life Expectancy
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: Public Health Statistics Yearbook, Sofia, 1999.1/ As of 1996, the age groups are: 0-17, 18-59, and 60+.Bulgaria: Population Dynamics and Life Expectancy
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: Public Health Statistics Yearbook, Sofia, 1999.1/ As of 1996, the age groups are: 0-17, 18-59, and 60+.Bulgaria: Population Dynamics and Life Expectancy
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: Public Health Statistics Yearbook, Sofia, 1999.1/ As of 1996, the age groups are: 0-17, 18-59, and 60+.
Bulgaria: International Comparison of the Age Structure of the Population, 1997
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: OECD Health Database, and Bulgarian National Statistical Institute.1/ Data for Bulgaria refer to 1998.Bulgaria: International Comparison of the Age Structure of the Population, 1997
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: OECD Health Database, and Bulgarian National Statistical Institute.1/ Data for Bulgaria refer to 1998.Bulgaria: International Comparison of the Age Structure of the Population, 1997
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: OECD Health Database, and Bulgarian National Statistical Institute.1/ Data for Bulgaria refer to 1998.Bulgaria: Pension Indicators
Annual Averages, in thousands
Women 55 and over; men 60 and over
Bulgaria: Pension Indicators
Annual Averages, in thousands
| 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | |
|---|---|---|---|---|---|---|---|
| Population | 8,472 | 8,444 | 8,406 | 8,363 | 8,312 | 8,257 | 8,211 |
| People in pensionable age 1/ | 2,017 | 2,030 | 2,038 | 2,040 | 2,037 | 2,031 | 2,026 |
| Employment | 3,222 | 3,242 | 3,282 | 3,286 | 3,157 | 3,168 | 2,811 |
| Number of insured persons | 3,030 | 2,940 | 3,092 | 3,114 | 3,210 | 3,084 | 2,863 |
| Number pensions | 2,492 | 2,481 | 2,465 | 2,442 | 2,433 | 2,436 | 2,430 |
| Number of pensioners | 2,442 | 2,432 | 2,417 | 2,395 | 2,387 | 2,390 | 2,384 |
| in percent | |||||||
| People in pensionable age/population | 23.8 | 24.0 | 24.2 | 24.4 | 24.5 | 24.6 | 24.7 |
| Number of pensioners/number of pensions | 98.0 | 98.0 | 98.1 | 98.1 | 98.1 | 98.1 | 98.1 |
| Number of pensioners/population | 28.8 | 28.8 | 28.7 | 28.6 | 28.7 | 28.9 | 29.0 |
| Number of pensioners/employed | 75.8 | 75.0 | 73.6 | 72.9 | 75.6 | 75.4 | 84.8 |
| Number of pensions/insured people | 82.2 | 84.4 | 79.7 | 78.4 | 75.8 | 79.0 | 84.8 |
Women 55 and over; men 60 and over
Bulgaria: Pension Indicators
Annual Averages, in thousands
| 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | |
|---|---|---|---|---|---|---|---|
| Population | 8,472 | 8,444 | 8,406 | 8,363 | 8,312 | 8,257 | 8,211 |
| People in pensionable age 1/ | 2,017 | 2,030 | 2,038 | 2,040 | 2,037 | 2,031 | 2,026 |
| Employment | 3,222 | 3,242 | 3,282 | 3,286 | 3,157 | 3,168 | 2,811 |
| Number of insured persons | 3,030 | 2,940 | 3,092 | 3,114 | 3,210 | 3,084 | 2,863 |
| Number pensions | 2,492 | 2,481 | 2,465 | 2,442 | 2,433 | 2,436 | 2,430 |
| Number of pensioners | 2,442 | 2,432 | 2,417 | 2,395 | 2,387 | 2,390 | 2,384 |
| in percent | |||||||
| People in pensionable age/population | 23.8 | 24.0 | 24.2 | 24.4 | 24.5 | 24.6 | 24.7 |
| Number of pensioners/number of pensions | 98.0 | 98.0 | 98.1 | 98.1 | 98.1 | 98.1 | 98.1 |
| Number of pensioners/population | 28.8 | 28.8 | 28.7 | 28.6 | 28.7 | 28.9 | 29.0 |
| Number of pensioners/employed | 75.8 | 75.0 | 73.6 | 72.9 | 75.6 | 75.4 | 84.8 |
| Number of pensions/insured people | 82.2 | 84.4 | 79.7 | 78.4 | 75.8 | 79.0 | 84.8 |
Women 55 and over; men 60 and over
9. The collection of social insurance revenue was hampered by adverse incentives, a shrinking revenue base, and weaknesses in social security administration. The adverse incentives stemmed from high total social contribution rates (47 percent for the median employee, and up to 61 percent for some employees), widely varying pension contribution rates (from 22 percent for the self-employed to 47 percent for some groups), and a weak link between contributions and benefits. The reduction in the contribution base in part reflected the transformation of the economy: cuts in budgetary employment and the privatization and restructuring of state enterprises shrank the NSSI’s traditional contribution base. However, the emergence of social insurance arrears and very weak collections from the private sector (which toward the end of the 1990s accounted for around 60 percent of economic activity, yet only contributed around 10 percent of social insurance revenues) indicate that evasion was common and enforcement lacking. All in all, social insurance revenue fell from the pre-transition level of 11 percent of GDP to a low point of 7 percent of GDP in 1996–7.
10. A surge in early retirement and disability pensions further eroded the pension system’s long-term viability. The onset of the transition in Bulgaria brought a sharp increase in early retirement (Figure 5). On the one hand, this contributed to alleviating social tensions and improving the finances of the state enterprise sector by helping to reduce overstaffing while keeping measured unemployment in check.3 However, it lowered the average effective retirement age to just 56 years in 1998, imposing substantial costs in terms of unfunded pension liabilities. Several factors accounted for the surge in early retirement, including the mushrooming of early retirement schemes (in 1998, there were 11 different instances of laws and regulations giving rise to early retirement rights); the low opportunity cost of retiring (retirees continued to receive their pensions if they worked); and weaknesses in the control and enforcement of pension rights. In 1998, about a third of all new pensioners retired early, and one out of every four pensioners was an early retiree. The early retirement schemes reduced the number of social insurance contributors, further worsening the dependency ratios and contributing to the deterioration in the replacement ratio.4 A similar issue arose on a smaller scale in disability pensions: the number of people receiving disability pensions increased sharply in the mid-1990s (Figure 6).
Bulgaria: Indicators of the Pension System
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI; Pension Yearbook, 1998; and Fund staff estimates.Bulgaria: Indicators of the Pension System
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI; Pension Yearbook, 1998; and Fund staff estimates.Bulgaria: Indicators of the Pension System
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI; Pension Yearbook, 1998; and Fund staff estimates.
Bulgaria: Age Composition of Pensioners, 1998
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI; Pension Yearbook, 1998; Public Health Statistics Yearbook, 1998; and Fund staff calculations.Bulgaria: Age Composition of Pensioners, 1998
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI; Pension Yearbook, 1998; Public Health Statistics Yearbook, 1998; and Fund staff calculations.Bulgaria: Age Composition of Pensioners, 1998
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI; Pension Yearbook, 1998; Public Health Statistics Yearbook, 1998; and Fund staff calculations.11. The lack of indexation helped to keep pension expenditure under some control. While the indexation of pensions has been a major reason for rapidly increasing pension expenditures in a number of other transition countries, in Bulgaria formal indexation has not been in use even at the time of high inflation. Instead, the Council of Ministers has issued decrees to implement adjustments to nominal pensions. These decrees came monthly in February 1997 at the height of hyperinflation, with the intervals becoming longer (semiannual from 1998) as inflation was brought under control. This system enabled the authorities to limit pension expenditures to financeable levels (under 10 percent of GDP throughout the 1990s).
12. However, the real value of pensions eroded, creating social and political pressures for reform. In 1999, the average monthly pension stood at the equivalent of US$35, having declined 60 percent in real terms during the decade. As pensions were typically the only source of old-age income, this erosion created strong social and political pressures for pension expenditures and highlighted the long-term need for funded components in the pension system.
Pension reform
13. The authorities’ policy response to the accumulated problems of the old pension system was a decision to create a modern three-pillar pension system. The new pension system was launched from the beginning of 2000, following extensive preparatory work and a public awareness campaign. The new system being put in place is a modem three-pillar system, consisting of a strengthened version of the existing PAYG system (first pillar) and two fully funded pillars (a mandatory public pillar for new entrants, and a voluntary private one).5 The legislative basis for strengthening the first pillar (the Social Insurance Code) was adopted by parliament last year, and phased implementation started from the beginning of 2000. The universal second pillar will be launched in 2002, and preparations have already started. The legislative basis for the third pillar was adopted last year, paving the way for the creation of voluntary defined-contribution funds. Box 1 provides explanations of the concepts and terminology associated with the reform.
14. The pension reform involves a significant strengthening of the existing PAYG system. Key changes include substantially reducing early retirement categories, raising the retirement age (although parliament decided not to raise the regular and early retirement ages to the extent originally envisaged), and establishing a link between contributions and benefits. The reforms will lower the dependency ratio through a 6-month increase in the retirement age every year until it reaches 60 for women and 63 for men. The base for determining pension benefits now depends on the total length of contributing service and the actual amount of contributions paid, significantly improving the incentives to contribute on actual labor income. Moreover, the scope of early retirement for special groups has been further narrowed.6 Early retirement has been made available on a universal basis, provided the sum of contributing years and age—referred to as points—exceeds certain thresholds (98 points for men and 88 points for women this year).7 Pensions will be increased every year starting from July 1, 2001 based on a backward-looking “Swiss formula” depending on the previous year’s increase in social insurance revenues and the consumer price index, but will be limited to under four times the minimum old-age pension through end-2003. Also, from the beginning of 2002, those born after January 1, 1960 will be obliged to participate in universal and/or professional pension funds, established and managed by licensed pension insurance companies.
Bulgarian Pension Reform: Technical Definitions
Indicators of demographic pressure
Old-age dependency ratio: the number of people over 64 divided by total population
Total dependency ratio: the number of people under 15 and over 64 divided by total population
System dependency ratio: the number of pensioners divided by the number of contributors
Categorization of pension schemes by source of financing
Pay-as-you-go (PAYG) pension schemes – such as the pre-reform pension system in Bulgaria – finance pension payments from current payroll taxes
Funded pension schemes finance current pension obligations using returns from accumulated assets
Categorization of pension schemes by benefit payment formula
Defined benefit plans (for PAYG or partially funded schemes) define benefits received based on work or contribution history and realized returns on invested funds
Defined contribution plans (for fully funded schemes) pay benefits—typically, an annuity—depending on returns earned on lifetime contributions
The three pillars of the reformed Bulgarian pension system from January 1, 2000
Pillar I: existing PAYG scheme with modified parameters and almost unified contribution rates.
Pillar II: now encompasses occupational and special category contributors only; from 2002 onward, envisaged to be a universal, mandatory, privately managed, and defined contribution scheme
Pillar III: a new voluntary, private, defined contribution pension scheme
15. The reform includes substantial institution building. It created two funded pillars, one of which (the second pillar) at present remains restricted to certain professions and will only become universal from the beginning of 2002. A regulatory framework was created through the secondary legislation related to pillar III, stipulating the rules for setting up and managing universal or professional pension funds, formulating investment restrictions, the supervisory framework, the required management structure, and the minimum number of subscribers necessary for operation. In addition, the NSSI has been implementing a comprehensive social insurance administration modernization project with World Bank assistance. This project is aimed at improving financial monitoring and enforcement through enhanced information technology and institutional capacity building. The implementation of this project is proceeding on track, and major advances have been made in computerization, the tracking of revenue and expenditure flows, and in audit and enforcement procedures.
16. The reform involves compromises to limit up-front costs, but has beneficial effect on savings and factor markets. While the original plan foresaw uniform contribution rates in the reformed PAYG pillar, employers of special category workers are now required to pay 3 extra percentage points to the first pillar to limit the decline in revenues, while the original excess in the contributions paid on their behalf (12 percent for category I and 7 percent for category II labor) has been assigned to individual funded accounts in pillar II. Further transitory costs will depend on the take-up rate of the third pillar, and of the universal second pillar from 2002. This may further reduce the contributions to the first pillar, in part through evasion. On the upside, the new system may have other positive effects, including improved domestic savings (Box 2).
The positive impact of pension reforms
Pension reforms can promote of domestic savings, thus boosting investment and growth. While the increase in domestic savings is difficult to estimate and tends to materialize over a longer period of time, several studies (e.g., IMF Occasional Paper #153) confirm that well-designed pension reforms can significantly boost domestic savings via two main channels:
Alleviating the cashflow imbalances of the PAYG component through a reduction in the replacement ratio raises public savings, with typically only a partial Ricardian offset on private savings.
Adding a funded component to the pension system enhances contribution compliance and has a positive net effect on private savings in the long run. This can be offset in part by the deficit of the PAYG component opened up during the transition phase, making transition rules governing the reallocation of contributions among the pillars of the pension system critically important.
Pension reforms also hold the promise of improving the efficiency of labor and capital markets. The high social contribution rates distort the labor market by raising the cost of labor. Since phasing in funded components of the pension system leads to the replacement of a distortive tax by saving explicitly linked to income level in retirement, labor costs for the employer will decline, boosting employment. As for capital markets, the rapidly growing funds seeking profitable investment opportunities are likely to foster the widening and deepening of financial markets.
Fiscal impact of pension reform
17. The pension reform turned part of the accumulated unfunded pension liabilities into explicit debt. The need to find resources to pay existing obligations while not using up all current contributions constrains the speed of moving toward a funded system and limits the scope for reducing the high social contribution rates. While the actual fiscal impact of the complex reforms now set in motion can only be projected with a significant margin of error, these considerations underline the crucial importance of being prepared to adjust parameters of the pension system, and of persevering with efforts to substantially broaden the social contribution base.
18. The method of financing additional costs associated with the pension reform determines how the burden is shared by current and future working generations. If the additional costs are covered by the current general government budget, the working generation has to both finance the pensions of retirees and set aside funds for its own retirement. To the extent that pension reform leads to a deficit financed through additional borrowing, or by privatization receipts, the burden is either placed on future generations in the form of additional debt or borne in the form of reducing the asset holdings of the state. Bulgaria used a combination of these financing methods, spreading the burden among generations.
19. While the strengthened PAYG system is designed to be sustainable over the long run, there will be initial deficits. Since the working population must finance the ongoing costs of the PAYG scheme while also contributing to accumulate savings to fund their own future retirement, the overhaul of the pension system comes at the price of up-front revenue losses for pillar I. This reflects the transition problem encountered by all pension reforms which phase in funded components while maintaining a PAYG scheme. However, this impact will be more than offset over time by lower expenditure as the tighter retirement parameters (higher retirement age and reduced scope for early retirement) take effect.
20. This is demonstrated by the results of calculations using the NSSI model of the PAYG component based on an extensive set of detailed demographic and macroeconomic assumptions. This annual model was utilized during the design stage of pension reform in Bulgaria.8 The calculations presented are based on a set of assumptions and the version of the model dating from December 1999. The results of quantitative calculations strongly support the case for pension reform. Hypothetical NSSI deficit estimates resulting from continued operation of the old system are compared with the PAYG component of the new three-pillar system over a horizon of 50 years. To facilitate the comparison, the underlying assumptions on demographic and macroeconomic developments, as well as on the discount factor and the costs of financing the cumulative deficits are identical in the two scenarios.9 Table 2 presents the summary results for the first pillar, comparing the long-term discounted sum of the deficits under the old and the new systems as a percentage of 1999 GDP. Figure 7 presents the evolution of the deficits over time, demonstrating that up-front costs are substantial, but declining, and that pillar I can attain a surplus from the mid-2020s. Both convey a very convincing message: (i) in the absence of reforms, the cumulative deficits would have rapidly placed an unsustainable burden on the budget, and (ii) the reforms eliminate the bulk of the deficits. While no projection was performed to quantify the projected effect of the Unified Revenue Agency, the deficits remaining under the new system appear sufficiently small to be within the range that can be eliminated through improved revenue collection or modified eligibility or indexation rules. It is important to note that while the precise quantitative results are quite sensitive to the underlying assumptions, the firm conclusion that the reformed pension system is a distinct and substantial improvement remains valid for a wide range of assumptions.
Bulgaria: Net Present Value of Cumulative Deficits as Percent of 1999 GDP
Bulgaria: Net Present Value of Cumulative Deficits as Percent of 1999 GDP
| Old System | New System | |
|---|---|---|
| Without Financing Costs | 70 | 7 |
| With Financing Costs | 160 | 24 |
Bulgaria: Net Present Value of Cumulative Deficits as Percent of 1999 GDP
| Old System | New System | |
|---|---|---|
| Without Financing Costs | 70 | 7 |
| With Financing Costs | 160 | 24 |
Bulgaria: Projected Deficits of Pillar I With and Without Reform
(In percent of GDP)
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI, long-term projection model.Bulgaria: Projected Deficits of Pillar I With and Without Reform
(In percent of GDP)
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI, long-term projection model.Bulgaria: Projected Deficits of Pillar I With and Without Reform
(In percent of GDP)
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: NSSI, long-term projection model.Key remaining issues
21. Despite good progress, continued efforts are needed to ensure the long-run viability of the PAYG system. The various steps of the reform should be implemented effectively and in a timely manner, and complemented by a combination of enhanced revenue collection and further restrictions on entitlements (including a reduction in the length of the decade-long transition period) to eliminate the remaining cumulative deficit of the first pillar. Improvements in revenue collection can be achieved not only through the implementation of the Unified Revenue Agency abut also through improved tax policy design. In particular, consideration should be given to correcting the existing distortion in favor of the self-employed and those working without a formal labor contract (e.g., consultants and other service providers). Participants in this group of contributors can choose their contribution base, the type of social risks they want to insure against, and are entitled to a lower contribution rate. The ranks of these contributors are swelling rapidly, and most of them declare income to the NSSI at or under the legal floor of two minimum wages, with nothing in the pension reform obliging them to change their behavior. As a result, the private sector which accounts for the bulk of the employment in the economy pays just a tenth of all social contributions. This precludes a substantial reduction in the high contribution rates for those formally employed with a labor contract, which is distortive for the labor market and has a deleterious effect on incentives to contribute.
22. Much work is also ahead to get the funded components of the pension system off the ground. The key future tasks in the implementation of the second and third pillars include the following:
The government needs to strengthen its supervisory and regulatory capacity and put in place a system to record the contributions for the pension system as well as strong safeguards to ensure that an independent depository holds the accumulated assets.
To ensure the efficiency of supervisory and regulatory activity, accounting, reporting, and other rules must be clear and strictly enforced.
If government guarantees are provided for the various pillars of the pension system, they need to be explicit, and designed in a manner that minimizes moral hazard and adverse selection of participants.
The investment restrictions on voluntary private pension funds (pillar III) severely constrain their earning capacity. Thus, they will need to be reviewed and substantially eased, in line with improvements in pension fund supervision.
Operating costs of funded schemes must be minimized, including through centralized collection of second and third pillar contributions. Competition, restrictions on investments, and rules on opting in and out of the funded pillars will affect these costs. Administrative costs can also seriously dent the returns to such schemes.10
Public support needs to be maintained by collaborating with social partners, providing information on a timely basis, and implementing reform at a steady pace.
A reform of life insurance regulation is needed to determine the role of the state in this area, including the rules and incentives for providing the annuity portion of old age pensions, and of survivors’ and disability benefits.
C. Reform of the Health Care System
The old health care system
23. The Bulgarian health system inherited from the past was a highly centralized and entirely budget-financed system which used resources increasingly inefficiently. The system was based on centralized public provision of health care which was financed from general budget revenues. It was heavily biased toward hospital care, with an oversupply of hospital beds but limited resources available for basic and preventive care (Figure 8). Although this system helped to improve public health indicators during the 1960s and 1970s, it lacked the flexibility to respond to an increase in the incidence of chronic diseases since the mid-1980s. Moreover, it was not able to reduce infant mortality rates to levels prevalent in developed nations or in the leading transition countries. The system also failed to provide appropriate incentives for health care providers to improve their services and for the population to adopt a healthier lifestyle. Finally, it could not adjust in an orderly manner to the substantial reduction in real resources devoted to health care during the transition period. Consequently, despite the professional integrity of most health care providers, its guarantee of free and accessible healthcare for all citizens rang increasingly hollow. The health care infrastructure deteriorated, the quality of care declined, and significant side-payments became the norm.
International Comparisons of Health Indicators, 1995
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source:1/ DEU, FRA and UKR data refer to 1993; HUN and POL to 1994.2/ DEU and UKR data refer to 1992; POL, 1993; HUN and ROM, to 1994.3/ Data refer to 1996.International Comparisons of Health Indicators, 1995
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source:1/ DEU, FRA and UKR data refer to 1993; HUN and POL to 1994.2/ DEU and UKR data refer to 1992; POL, 1993; HUN and ROM, to 1994.3/ Data refer to 1996.International Comparisons of Health Indicators, 1995
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source:1/ DEU, FRA and UKR data refer to 1993; HUN and POL to 1994.2/ DEU and UKR data refer to 1992; POL, 1993; HUN and ROM, to 1994.3/ Data refer to 1996.24. Limited reforms in the 1990s aimed primarily at decentralizing health care provision could not arrest the deteriorating trend. Reform measures through 1998 included allowing the provision of private health services, establishing medical associations, privatizing most pharmacies, and devolving the responsibility for selected health care services to municipalities.11 These measures led to a marginal reduction in the number of medical personnel per 10,000 people (Figure 9). However, they could not prevent the emergence of a compartmentalized infrastructure where quality health care was mainly available in the parallel systems maintained and separately financed by ministries.12 Coordination between the parallel systems, as well as between the central government and municipalities, was weak. The outcome was a health system that employed an inordinately high number of health care providers on a per capita basis, but failed to deliver quality health care at an affordable price to the population.13
Bulgaria: Medical Personnel
(per 10,000 population)
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: Public Health Statistics Yearbook 1998, Ministry of Health.Bulgaria: Medical Personnel
(per 10,000 population)
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: Public Health Statistics Yearbook 1998, Ministry of Health.Bulgaria: Medical Personnel
(per 10,000 population)
Citation: IMF Staff Country Reports 2000, 064; 10.5089/9781451804447.002.A002
Source: Public Health Statistics Yearbook 1998, Ministry of Health.Health reform
25. Bulgaria’s health reform aims at a suitable mix of market and non-market mechanisms. While the history of health care provision and financing in Bulgaria has provided ample evidence of the inefficiencies of government-provided care, adopting pure market solutions would not be desirable either: there are externalities inherent in health care, social norms prescribe some access for at least a basic benefit package regardless of the ability to pay, and asymmetric information between the parties to a health insurance contract leads to adverse selection and moral hazard. Moreover, some rationing of health services is inevitable, because the highest possible standards of health care would be prohibitively expensive, i.e., the budget constraint would become binding well before the level of supply reached a saturation point. International experience also shows that well-functioning health care systems all involve a mix of market and non-market mechanisms (see Kornai, 1998, for a particularly comprehensive survey).
26. Bulgaria’s health care reform comprises an overhaul of financing, a reorganization of primary health care provision, and a rationalization of hospital care. The first element was implemented through the 1998 adoption of the Health Insurance Law which led to the creation of the Health Insurance Fund (HIF) from mid-1999. The second element is the essence of the next round of health care reforms to be launched from mid-2000, while the third part of the reform will be implemented from mid-2001. The phased implementation is in line with the objective of designing and implementing an improved system of health care provision and financing in a manner that minimizes disruptions to the existing system during the interim period. It also ties in with the phased implementation of the unified revenue agency, which was designed to ensure that the necessary revenue base will be available to finance social reforms.
27. The Health Insurance Act fundamentally reforms the financing of health care in Bulgaria through universal compulsory insurance. This Act created the Health Insurance Fund as a statutory insurer and bulk purchaser of health care services. The HIF is answerable to parliament which approves its annual budget. Its revenues accrue from payroll contributions presently set at 6 percent of gross wages (divided equally between the employee and the employer), and collected by the NSSI since mid-1999.14 The HIF has 28 regional offices, and is currently building up its information technology, monitoring, audit, and health care administration capacity. By April 2000, it will have signed a national framework agreement with the representative body of primary health care providers, and will soon conclude bilateral service provision contracts with doctors prior to the launch of the new primary and specialized health care system from July 1, 2000. Hospital care, population health services (epidemiology, blood transfusion centers, etc.), expensive treatments (e.g., cancer therapy), as well as capital investment and health education will continue to be funded by the budget. Starting during 2000, the intention is to provide access to supplementary insurance for health care that is excluded from the basic benefit package. In parallel with the implementation of the HIF-financed health care for basic services, the preparations for the mid-2001 launch of similarly financed hospital care will begin. By mid-2000, hospitals will have to be incorporated as separate corporate entities. By end-2000, details of providing hospital care (including a detailed pricing mechanism covering all services) will be proposed to hospitals, followed by the signing of a framework contract and bilateral agreements with hospitals, both public and private.
28. The health reform also aims at revamping primary health care and rationalizing the hospital network. In addition to retraining family doctors and establishing a clearer division between primary and other health care, it is expected that the markedly changed incentives and the increased level of competition among primary health providers will have a salutary effect on the quality of health care and the willingness to invest in diagnostic and other medical instruments. A growing share of health care services is expected to be provided by private sector entities as the reform process unfolds. Moreover, a number of underutilized hospitals, and those in a very poor state of repair, will be closed or downsized, including by municipalities, in response to the imposition of hard budget constraints.
29. Health care reforms were formulated with a view to ensuring a limited fiscal impact. The basic health care package (comprising primarily recurrent clinical health care) has been defined within the revenue capacity of the HIF. If all the assumptions materialize as envisaged (see next paragraph), health-related budgetary expenditures will still rise in 2000, but to a limited extent: the 2000 budget envisages an increase in the share of health expenditures in GDP of approximately ½ percent, reflecting to a large extent non-recurrent institution-building expenses and the need to own up to years of delayed capital investments. Limited unforeseen excesses over this level can be accommodated using special contingency funds set aside in the 2000 budget in part for this purpose.
30. Turning to the remaining challenges, the key ones are financing hospital care, avoiding a marked increase in health care costs, and providing supplementary insurance. Adequate long-term financing for quality hospital care needs to be found. At present, health contribution rates are well below those in leading transition economies (6 percent versus 13.5 percent in the Czech Republic and 19.3 percent in Hungary), but the already very high overall social contribution rates leave little room for increase. Economic growth and enhanced revenue collection may allow a higher level of spending on health care than in previous years, but continued significant central government (and municipal) transfers are likely to be required for the health sector. Another key issue is whether the HIF will have the capacity to strictly control the costs of health care provision. Part of the reason for escalating costs is the characteristic of health care that technological progress in this sector does not go hand in hand with savings in labor or costs. In fact, the typical medical advance delivers a higher survival rate or improved comfort levels for patients at higher levels of inputs. International experience is mixed with regard to the ability of monopsonist buyers to contain surges in health care costs, an outcome that must be avoided to ensure the success of health reform. Finally, problems may arise owing to the lack of insurers ready to provide complementary health insurance. Experience in advanced transition economies has shown that private insurers may not be able or willing to rapidly fill in this gap at affordable insurance rates.
D. Concluding Remarks
31. Bulgaria’s pension and health reforms can also be evaluated against a broader set of requirements. So far in this paper the discussion of pension and health reforms has been primarily fiscal in nature. However, it may be of some interest to review the pension and health reforms in a broader context, utilizing János Kornai’s principles of social reforms as the framework for assessment (Box 3).
János Kornai’s Principles of Social Reform
These principles describe the desirable properties of social reforms. The first two are ethical postulates, the following five describe the necessary attributes of the institutions and coordination mechanisms created by the reforms, while the final two refer to long-term aspects of allocational efficiency.
(1) Individual sovereignty—the need to enhance the individual’s right to choose.
(2) Social solidarity—the need to support the poor and disadvantaged.
(3) Competition—let coordination mechanisms and public and private provision of services compete.
(4) Incentives—the reformed system must provide incentives for both the providers and the users of services to use resources efficiently.
(5) Transforming the role of the state—restrict intervention to areas of market failure where the state can improve on the outcome.
(6) Transparency—design and implement reform in open interaction with the public, and make the cost of providing additional services clear to taxpayers.
(7) Managing the transition—allow sufficient time for thorough preparation (including for consensus-building) and effective implementation; provide temporary support to those adversely affected and unable to adapt rapidly but avoid permanently raising subsidies; and rectify emerging teething problems.
(8) Balanced growth—strike an optimal balance between allocating resources for welfare reforms and growth-enhancing investment.
(9) Fiscal sustainability—ensure continued financing of the state’s role enshrined in legislation.
32. If fully implemented, Bulgaria’s pension and health reforms appear to satisfy the main criteria of successful social reforms, as defined by Kornai. These reforms represented a clear break with the past by substantially improving the scope for individual sovereignty, while retaining aspects that embody social solidarity. Freedom of choice has dramatically increased in both areas, while the strong remaining redistributive elements in the first pillar of the pension system, together with the maintenance of the constitutional right to basic health care, show that reform design took into account the requirement to support the poorer segments of the population. Various forms of service providers are now free to compete for pension contributions and for the right to provide health care. The incentives to contribute to social insurance, and to adopt a healthier lifestyle have been strengthened because the financial cost of doing otherwise has increased, but also because the future rewards in terms of a higher pension or improved health care can be reasonably expected to materialize as a result of the reforms. One of the most important aspects of these reforms is the dramatic reduction in the role of the government, a shift as important in creating a well functioning market economy as privatization. Government will only play the role of the regulator and supervisor in two of the three pillars of the pension system, while health care provision is being transformed from a 100 percent state owned and run sector to one containing both private and public operators. The reform process holds the promise of substantially increasing transparency in areas that have been characterized by a lack of clear connection between inputs and outputs, unfair advantages provided to the selected few, and deals involving side-payments. The preparation of pension and health reform—involving consensus-building and public education campaigns—was thorough and extensive, and the initial experience in the implementation phase is positive.
REFERENCES
Benczur, P., 1999/8, Estimate of the changes in implicit public debt implied by the Hungarian pension system (in Hungarian). National Bank of Hungary Working Paper.
Brugiviani, A., R., 2000, Disney, European Pensions: An Appeal for Reform. Pension Schemes that Europe Can Really Afford. A Report by the European Roundtable of Industrialists, Brussels, Belgium, January.
Bulgaria— Recent Economic Developments and Statistical Appendix. SM/99/30 Washington, February 1999.
Cangiano, M., C. Cottarelli, L. Cubeddu., 1998, “Pension Developments and Reforms In Transition Economies”. IMF Working Paper 98/151 (Washington: International Monetary Fund).
Costello-Branco, M., 1998, “Pension Reform in the Baltics, Russia, and Other Countries of the Former Soviet Union (BRO)”, IMF Working Paper 98/11 (Washington: International Monetary Fund).
Healy, J., H. Hinkov, S. Koulaksuzov, 1999, I. Semerdjiev: Health Care Systems in Transition Copenhagen: European Observatory on Health Care Systems, WHO Regional Office for Europe.
Kornai, J., 1998, On Health Care Reform (in Hungarian). Budapest, KJK.
Laursen, T., 2000, “Pension System Viability and Reform Alternatives in the Czech Republic”, IMF Working Paper 00/16 (Washington: International Monetary Fund).
Mackenzie, G. A., P. Gerson, A., 1997, Cuevas: Pension Regimes and Saving. IMF Occasional Paper No. 153, (Washington: International Monetary Fund).
Ministry of Health, Bulgaria: Public Health Statistics Yearbook. National Center of Health Informatics, Sofia 1999, ISSN 1310-2931
Murthi, M., J. M. Orszag, and P. R. Orszag, 1999, Administrative Costs under a Decentralized Approach to Individual Accounts: Lessons from the UK. Paper presented at the Conference on New Ideas about Old Age Security, September. The World Bank, Washington DC.
National Social Security Institute, Bulgaria: Statistical Yearbook 1999.
National Social Security Institute: 1998 Yearbook of Pension Statistics, Sofia, 1999
Pension Reform Working Group, 1999, Bulgaria: Bulgarian Pension Reform Strategy. Working Paper, July.
Rutkowski, M., Bringing Hope Back: Pension Reforms in Transition Economies. Conference paper in the Ten Years After Conference on Transition Economies, October 15-16, 1999 in Warsaw, Poland.
Staines, V. S., 1999, A Health Sector Strategy for the European and Central Asia Region Human Development Network, Health, Nutrition and Population Series. The World Bank, Washington, DC.
Prepared by Balázs Horváth.
Special category workers were those who worked in occupations deemed especially strenuous or dangerous, including miners, sailors, and workers in the armed services and the police.
This use of the pension system as a tool of social policy has been quite common in transition economies.
The replacement ratio grew steadily during the pre-transition era. In 1975 the average pension amounted to 33.3 percent of the average wage; in 1980 the replacement ratio grew to 37.4 percent, and in 1985 to 43.8 percent, remaining around that level through 1989.
Similar reformed multipillar pension schemes exist today in Argentina, Australia, Colombia, Chile, Denmark, El Salvador, Hungary, Kazakstan, Mexico, the Netherlands, Peru, Poland, Slovenia, Sweden, Switzerland, UK, and Uruguay.
Military personnel will continue to retire after 25 years of service, and three (four) years of service in work category I (II) shall be counted as 5 years of pensionable service. Part of the social contributions for these workers is allocated to build up additional funded pension rights, together with those in occupational retirement schemes (e.g. teachers).
The required number of points will increase by 1 a year until reaching 100 for men and 90 for women.
The model was designed and calibrated by John Wilkins, working with NSSI experts and members of the Pension Reform Group. The calculations were performed by Gabriella Stoyanova of the Agency of Economic and Analysis and Forecasting.
For simplicity, 3 percent inflation and 2 percent real discount rate per annum was assumed; financing costs were set at 5 percent per annum through 2048.
Administrative costs of funded schemes can be a significant problem, especially in the accumulation phase. Murthi, Orszag, and Orszag 1999) calculated that in the decentralized U.K. system up to 40 percent of the value of an individual retirement account could be dissipated through market transaction fees and charges.
In 1998, municipalities provided 55 percent of state funding to the health care system. They own and finance polyclinics, as well as some general and specialized hospitals.
In 1999, the Ministries of Defense, Internal Affairs, and Transport each owned several hospitals and maintained parallel hygiene and epidemiology services.
Almost two thirds of the respondents to a 1995 Ministry of Health survey on the quality of primary and secondary health care regarded it as “bad” or “very bad”.
The budget pays health insurance contributions for pensioners, the unemployed, and others outside the labor force.
