Social Sector Reform in Transition Countries

Contributor Notes

Author’s E-Mail Address:;

During the transition process, many existing social sector institutions and policies were significantly eroded and their underlying character changed. As a result, they often do not redistribute to the poorest, nor generally serve the role of facilitating economic change. Social sector reforms have therefore become necessary for reasons of social welfare as well as economic growth. The analysis of eleven transition countries—comprising some of the most advanced as well as some of the poorest transition economies—shows that almost all countries have started to undertake reforms; however, their individual efforts vary. Reform does not only stand for cutting back, but also requires in some cases a building up and in others a redesign of social safety nets; it needs to address insurance issues, budgetary transfer programs, the performance of the health and education sector, as well as the labor market regime and the approach to tax administration.


During the transition process, many existing social sector institutions and policies were significantly eroded and their underlying character changed. As a result, they often do not redistribute to the poorest, nor generally serve the role of facilitating economic change. Social sector reforms have therefore become necessary for reasons of social welfare as well as economic growth. The analysis of eleven transition countries—comprising some of the most advanced as well as some of the poorest transition economies—shows that almost all countries have started to undertake reforms; however, their individual efforts vary. Reform does not only stand for cutting back, but also requires in some cases a building up and in others a redesign of social safety nets; it needs to address insurance issues, budgetary transfer programs, the performance of the health and education sector, as well as the labor market regime and the approach to tax administration.

I. Social issues and the IMF

Although only indirectly linked to its mandate, the Fund regards issues of social welfare as fundamental. Policy objectives such as macroeconomic stability, external viability, and medium-term sustainability are not ends in themselves, but are means to ultimately improve the social welfare in countries seeking the Fund’s advice. Sound macroeconomic policies, growth-enhancing structural reforms, and good social policies are the conditions for sustainable growth that generate higher levels of employment and real income. Growth that is extremely unequal, benefiting only a few and inhibiting widespread social improvement, has proven unsustainable and is therefore not satisfactory.2

Beginning in the mid-80s, significant changes have taken place in the Fund’s perspective on these issues and the weight that it attaches to the area of social policy.3 During these years there has been greater recognition of the impact macro-adjustment policies can have on the social sector and on social development. As a consequence, the Fund has taken a progressively more active stance on social policies to ensure that they are incorporated into Fund-supported programs. At the same time, the Fund has intensified its collaboration with the World Bank and other international agencies that have social policy expertise.

In general, what have we learned? It is has become common ground that sustainable growth and macro stability are the essential engines to address poverty. Growth creates employment and thus the opportunity for the poor to raise their income and consumption levels. At the same time, it is also widely recognized that high inflation and overvalued exchange rates hit the poor the hardest, as their nominal fixed incomes quickly erode and they often depend on export agriculture.

Nevertheless, growth and macro stability by themselves are not sufficient. Macroeconomic adjustments and policies to promote economic growth often adversely affect some groups in society. These groups may include poor, but also non-poor, and while many poor benefit, some poor may lose. In other words, economic change creates social costs. The Fund, for its part, must consider these social costs when designing adjustment programs, and make sure that adequate social safety nets are in place to catch those who are at risk of falling below an acceptable minimum standard of living as a result of the adjustment process.

II. Key role of the government in the social sector

It is important to underscore that the government’s role in the social sector is a challenge faced by any government, whether from a transition, developing or developed country. Indeed, social policies play a vital role in mature economic systems, even under conditions of normal growth. First, the shifting dynamics of economic change may leave some behind, becoming unable to generate sufficient income in the market process at all. For those, social transfer programs (social welfare) must provide income support. Second, social insurance systems can facilitate sectoral changes, by mitigating the often painful adjustments forced upon individuals. Unemployment insurance schemes, for example, smooth the transition of workers who need to adapt to shifts in economic activity. While the labor market should be flexible enough to allow for such changes, unemployment benefits can support the process by ensuring that workers have the time and resources to search for adequate jobs in new sectors. With such social protection in place, the risks associated with economic change become acceptable to everyone. This creates the sufficient condition for the market process to unfold its dynamics.

Third, social policies in the form of public spending on education and health care are a crucial issue for the developing economies. Much of the spending in these sectors constitutes investment in human capital, which is a decisive element in growth oriented strategies. Fourth, social insurance arrangements such as public pensions schemes play an important role in the general management of social risks. Traditional family-based and other informal systems of old-age provision may prove less sustainable in modern societies even in the absence of population aging. For many developed countries, however, responding to demographic change has become one of today’s biggest social policy challenges, and it will require fundamental reforms on their part.

III. Challenges for social policy in transition economies

A. Looking Back: Main Social Sector Developments During the Transition

Turning to transition countries, the following general picture can be sketched out.4 Notwithstanding the considerable problems of measuring the actual developments on basis of the poor benchmark data provided by the former regimes, it is obvious to all that the transition process gave raise to major economic and social problems, as employment and income fell sharply, and income inequalities widened. Social indicators such as life expectancy and school enrollment deteriorated, and the incidence of poverty rose. Although these developments varied considerably across countries and population groups—some countries were much worse hit than others, and while many households suffered some prospered—the economic and social decline the transition countries went through was traumatic, often exceeding the magnitude of the crisis that has been experienced by developing countries.

In contrast to most developing countries, however, transition countries had a well developed social sector before the transition. Their social safety nets covered the typical social risks usually covered by social insurance plans and transfer programs in developed countries. Furthermore, large amounts of resources were devoted to health care and education. Yet, the institutional arrangements of the social sector—providing “cradle-to-grave” protection to the entire population—had been designed for a very different economic system. They were often not compatible with the incentive mechanisms of a market economy, and in particular were not prepared to cope with the enormous pressures that emerged with the transition from a plan to a market economy.

Not surprisingly therefore, the existing social sector institutions and policies were significantly eroded and affected the transition process. Existing insurance schemes and transfer programs often failed to protected the vulnerable from poverty as the real value of social transfers and benefits was eroded by inflation. The purchasing power of pensions became strictly dependent on the specificities of indexation mechanisms or political decisions to make ad hoc adjustments.5 While this might have worked in favor of pensioners in a few cases such as Poland, it often meant that pensioners received not enough to cover minimum subsistence needs, as happened, for example, in Russia.6 The real value of unemployment benefits was often small, difficult to obtain, and paid out with such long lags that unemployed persons did not even apply.

Rapid price increases during the transition also affected health care and education. Medical supplies and drugs became no longer affordable for some. At the same time, real spending on health care and education was often cut to levels, at which—in the absence of strong efficiency improvements in these sectors—the provision of basic services in health care and education was no longer guaranteed.7 Immunizations and enrollment rates in secondary education declined in many transition countries.

Further, the transformation of the economy created social risks that the existing safety nets had not been designed to address, both with regard to the size and nature of the risks. As the social insurance systems were then widely used in ways different from their intended purpose, they became remarkably deformed. For example, while under central planning, unemployment was a negligible phenomenon, unemployment schemes suddenly faced surging unemployment numbers caused by a general drop in economic activity as well as privatization and enterprise restructuring.8 To deal with the pressures of swelling unemployment, the problem was shifted to the public pension schemes, which had to absorb older workers through early retirement plans and by easing the eligibility to disability benefits. Very low average retirement ages, and excessively large shares of pensioners receiving disability benefits reflected such policies.9 As a result, large gaps emerged between the ratio of pensioners to contributors (system dependency ratio) and the true ratio of old to young in the population (demographic dependency ratio).

Many of these developments could have been prevented by better policies. However, whereas the mechanism of a pension scheme is relatively straightforward, and its use for nonpension purposes therefore an indication of poor policy, governments also often faced problems where clear cut policy solutions simply did not exist. For example, many governments tried to expand or newly introduce social health insurance plans to finance health care services. Yet, while these plans tried to rely on the market to set prices, under the prevailing market structures, physicians and hospitals could use their market power to set monopolistic prices. And in cases where the insurance plans were allowed to set service fees, they often saw the volume of services soaring. As a result, their financial deficits rose—some even went bankrupt and had to be bailed out by the government (e.g., in the Czech Republic).10

The central governments commonly tried to transfer the problem of mounting costs for health care and education by decentralizing schools and public hospitals—which before often had been part of the large state enterprises—to local governments.11 The local governments, however, were in no better fiscal position to manage the problem. They lacked the funds to support schools which featured small classes and many teachers, and health care facilities which traditionally had excess capacities of hospital beds and redundant health practitioners. In addition, health sector personnel pressed for higher wages to make up for inflation, and opposed the privatization of public health facilities to prevent large-scale lay-offs.12

As a result, high capacities for curative care, including many spa and recreation facilities, were maintained, leaving little resources for preventive medicine and primary health care. In the educational sector, utility costs and wages absorbed large amounts of the expenditure allocations, whereas spending on teaching materials and on the maintenance of the educational infrastructure gradually sank.

At the same time, governments granted numerous new privileges in addition to existing general subsidies for certain goods and food stuffs. In many countries such privileges entitled large segments of the population to reduced tariffs for energy, telephone, housing, communal services, and transportation.13 The low prices created incentives for overconsumption and misuse, thus further straining the budget and distorting resource allocation.

While the transition process increased the demand for social benefits on the one side, it undermined the availability of financing on the other. The drop in economic activity reduced the revenue base, which in combination with the need to meet rising expenditure obligations, triggered a vicious circle. As statutory tax and contribution rates were increased, more and more tax payers tried to avoid paying them altogether by moving into the informal sector. Aided by the fact that nascent tax administrations were still weak, and authorities often lacked the political will to enforce the rule of law (e.g., by allowing contribution arrears to accumulate), tax compliance faded. The size of the economies’ informal sector expanded rapidly,14 and many employers shifted large components of labor compensation out of the official wage, reporting only minimum wages for their employees. Thus, the tax base shrank even further, intensifying the upward pressure on tax rates.

As a result, the effective social contribution rates (contributions divided by the wage bill) averaged in many countries as low as half of the statutory rates.15 When expenditure arrears accumulated, the respective agencies frequently resorted to collecting taxes and social contributions in kind. In some cases, social funds became reliant on extensive barter operations under which benefit and contribution obligations would be “netted out” in various ways (e.g., Moldova).

In the health and education sector, the lack of general tax revenue prompted the search for new revenue resources, in particular in cases where sudden decentralization had left local governments responsible for the financing of schools and hospitals. Service providers were encouraged to charge user fees in order to generate their own revenue. For poor households, this often entailed heavy out-of-pocket expenses just to receive basic health care and education, since the fees were typically levied where they could be best enforced, regardless of ability-to-pay considerations.16 Besides, the more affluent were able to offer additional under-the-table payments to government-employed health practitioners. While this ensured short waiting periods and expensive high-technology services for them, it also diverted time and resources away from the basic treatments most important for the poor.

Despite all these shortcomings, in particular the lack of targeting benefits and social transfers, the social policy decisions made during the transition should perhaps not be dismissed as failure altogether. The meaning of “targeting”, for example, may appear in a different light when one takes a political economy perspective of transition. True, the social sector developments described above signify that substantial proportions of social expenditure were directed not towards households at the bottom of the income distribution but towards the middle class—clearly a mistargeting from the welfare perspective. Yet, while not necessarily preventing an increase in poverty, the granting of benefits, subsidies, and privileges to large segments of the population may still have mitigated increasing gaps in income and consumption. And this, in fact, may be seen by some as an overriding goal, since it may have helped to gather wide political support for painful but necessary reforms. (See Box 1)

On balance, however, it can be argued that the transition process not only aggravated social sector problems but also notably changed its characteristics. The result is daunting in several ways: warped social insurance schemes that do not redistribute to the poorest, nor generally serve the role of facilitating economic change, transfer programs, and education and health services that have not only become glaringly inefficient, but often even ineffective. On the expenditure side, the generosity of some social insurance benefits, the lack of targeting of transfers, the skewed allocation of resources in the health and education sector, and excessive privileges have inflated the cost of social protection, while still leaving the most vulnerable without sufficient support. On the revenue side, high statutory tax and contribution rates have created massive disincentives in the labor market, impeding the creation of more employment. The unsystematic levying of user fees not only has clouded the transparency of the revenue system, but obstructed the access of the poor to basic services.

It is important to note that while these development were rooted in the nature of the transition process itself, they were exacerbated by poor policy implementation. To let the real value of benefits erode for some, while providing generously for others, to keep vulnerable groups waiting for their monthly wages or transfers, while allowing certain industries to accumulate large tax arrears, were not inevitable consequences of transition, but the results of policy choices. As in other areas of reform, vested interests often stymied the implementation of reform oriented policies also in the social sector.

“Targeting” from the Political Economy Perspective—the Case of Poland

The lack of targeting of benefits and social transfers has been one of the main criticisms of social policies in transition economies. Yet, as indicated above, there might be a political economy argument for using insurance schemes and social transfer programs to improve the income and consumption situation of large shares of the population, regardless of their actual vulnerability to poverty. Along this line, Keane and Prasad (2000) have emphasized in a recent paper that in Poland, the social transfer mechanism may have played a critical role in maintaining social stability and in reducing political resistance to the radical structural reforms that facilitated the country’s subsequent strong growth performance.

In particular, by providing a welfare cushion for the large number of older workers, both through early retirement and relatively generous pensions, the Polish government was able to remove the potential obstacles to enterprise restructuring. This policy was expensive—pension expenditure more than doubled from about 7 percent of GDP in 1988 to 15 percent in 1993—and to a large extend may have been unduly beneficial for middle-class households. Indeed, the authors provide evidence that the overall targeting of social transfers to the least well-off could have been substantially improved in Poland. Yet, so goes the argument, mitigating the fall of income (and consumption) for the middle-class—a group that has a significant higher propensity to vote than lower income groups—helped to “buy” the political peace that was needed to undertake the reforms. From that perspective, the social transfers were “well-targeted”.

However, prudence is called for when using this line of argument as general advice for other transition economies. Admittedly, Poland managed to generate strong economic growth rates, and—as Keane and Prasad find in their study—at the same time was able to limit the rise in inequality usually observed during transition. This allows them to conclude that policies to mitigate a rise in inequality do not necessarily have to come at the expense of growth, as had been traditionally suggested (e.g., the Kuznet hypothesis), and that they may even be conducive to growth, as Keane and Prasad’s results suggest. Nonetheless, the issue remains complex and many other factors come in play when explaining countries’ growth performances. While the lack of targeting might not have hurt in the Polish case, this by no means supports a simplified thesis that inefficient social policies are beneficial for growth.

Also, as result of the lax and little targeted transfer policies, the amount of cash transfers to individuals has reached 19 percent of Poland’s GDP—the highest level in all transition countries. Yet, while Poland’s spending on social transfers exceeds that of the other transition economies considered for early EU accession, its per capita income is the second lowest among them. Without analyzing this constellation any further, it gives an indication that policies which buy political peace in the short run may give rise to political conflict in the future when fiscal consolidation becomes necessary.

B. State of Play: Current Reform Efforts All Over the Map

Even if one accepts that past policies presented the only politically feasible choice under the special circumstances of transition, this argument loses its appeal after a decade of transition has passed. In fact, many countries have already taken or are taking steps now to reverse some of the policies chosen during the initial phase of transition. They recognize that substantial reforms of the social sector, comprising insurance schemes, budgetary transfer programs, and health and education services, are imperative for reasons related to social welfare as well as economic growth. Now that market mechanisms have been established in many areas, it is important for social policies to play their proper role in a dynamic market economy. In this vein, governments also seek reform strategies to rationalize expenditure and reduce the high levels of payroll taxes. Yet such strategies do not only stand for cutting back, they also require to build up, and in some cases may involve redesigning social safety nets altogether.

What is the state of play? Just as the for the overall transition process, it can be said for the area of social reform that there are more and less advanced countries. Yet, a direct comparison or even a ranking among countries is difficult. The countries started from different positions and the level of magnitude of the social problems they are facing varies largely. Further, a country’s reform effort might not have the same intensity in every area of the social sector, but is contingent on the country’s particular circumstances. For example, in Lithuania, a country with long and cold winters, governments typically face strong political resistance to eliminate privileges in energy consumption, and thus may lag in this important area of reform. In the Czech Republic, a country where the adverse changes in demographic structure will set in only in about a decade, governments might be reluctant at this point to use their political capital on unpopular pension reform measures. There is no unambiguous way of weighing efforts in one area against those in another.

With this in mind, however, a random group of eleven countries—comprising some of the most advanced as well as some of the poorest transition economies—have been compared with respect to the social sector reforms that have been reported recently.17 While this exercise can not claim to provide the complete picture of social reform in transition countries, it illustrates well some of the common developments in this area, as well as points out country specific problems and differences in the approach to reform.

The country sample has been tentatively divided into three rough groups. Strong reform efforts are being undertaken by Bulgaria and Poland in several social policy areas. A second group of countries, including Moldova, Lithuania, Romania and Slovenia, covers the middle ground. In these countries, reforms are underway, but have often not been implemented to the desired extent, and thus will require significant additions and further new reforms in the near future. In a third group, comprising Albania, Bosnia Herzegovina, Czech Republic, Georgia, and Slovak Republic, reform efforts appear to be rather limited to date, although they are dearly needed.

Strong reforms underway: Bulgaria, Poland

With their comprehensive reform laws of 1999, Bulgaria and Poland have embarked on pension reforms which are among the most ambitious in the transition world.18 In both countries, the reform follows a similar pattern. A three-pillar pension model has been introduced which consists of a strengthened version of the existing mandatory PAYG system (first pillar) and two fully funded pillars, a mandatory public pillar for new entrants and a voluntary private one. To strengthen the PAYG pillar, both countries are implementing—or have already done do so in previous years—parametric reforms to counter the past developments that threatened their PAYG systems’ finances. The eligibility for early retirement and disability benefits is being tightened, and the regular pension ages are being raised. Changes in the pension formulae ensure that replacement rates remain at reasonable levels, and that the link between contributions and benefits is strengthened.

Indeed, these parametric reforms are an essential part of moving towards three pillar systems. The gradual move of the workforce into the new system implies that the anticipated gains from the new system will only materialize over time, while immediate budgetary costs will arise due to the reduction of contributions going into the PAYG pillar. The complementary parametric reforms have created expenditure savings that can at least partially offset these costs.

At the same time, both countries have made significant progress regarding their health insurance systems. In Bulgaria a newly created Health Insurance Fund has started to negotiate contracts for out-patient care delivery with competing private and public providers, and is gearing up to do the same for in-patient care. Similarly, in Poland, local autonomous health funds have been established which will separate buyers and sellers of health services and presage an internal market for public health services. Financing of the health funds is provided by compulsory insurance contributions deducted from personal income tax withholdings.

Bulgaria is also taking steps to improve its collection of tax and social insurance contributions. A unified revenue agency will be created that can best deal with the persisting problem of arrears by integrating the collection, audit, and enforcement of taxes and social insurance contributions. Poland’s tax administration, however, is still relatively weak, given its advanced position in the transition process.

At the same time, Bulgaria still needs to improve the targeting of its social spending related to heating and electricity prices. The government has decided to freeze the prices for households for another heating season, although this is a weakly targeted measure and it would be more cost-effective to protect poor households through higher direct assistance payments. The price-freeze creates all the well known incentive problems, and could also impede the restructuring of the energy sector.

Both countries have taken initial steps towards increasing the flexibility of the labor markets and improving the incentive structure. In Bulgaria, amendments to the Labor Code have been submitted that include provisions to make hiring, firing and working hours more flexible. In Poland, progress has been made in recent years in tightening eligibility for unemployment benefits, and a new law curtails the wide abuse of sickness benefits. However, a reform of Poland’s minimum wage, which is well above that of other countries in the region, and more flexible fixed-term contracts, could further improve the conditions for job creation.

Reforms underway, but further efforts needed: Bosnia Herzegovina, Moldova, Lithuania, Romania, Slovenia

Similarly to the pension reform in Bulgaria and Poland, Slovenia also introduced a new supplementary, voluntary, and fully funded pension scheme. However, the financial situation of the existing PAYG scheme has eroded by a combination of generous pension increases, a reduction in the contribution rate and adverse demographic trends. Although parametric reforms, such as raising the retirement age, are underway, they miss the objective of fully restoring financial viability. Further corrective measures will become necessary in the future.

Moldova passed pension laws in 1999 which are a crucial step towards improving its Social Fund’s (SF) finances. They include an increase in the retirement age, and the introduction of individual social security accounts. In addition, the SF’s collection responsibilities have been moved over to the State Tax Inspectorate in order to improve collection and limit the practice of in-kind contributions. However, it remains to be seen to what extent these steps will be sufficient to overcome the SF’s large financial imbalances and its excessive reliance on barter operations.19

Lithuania has sought to improve its State Social Insurance Fund’s financial position by realizing expenditure savings associated with an accelerated increase in the pension age, a revised inflation adjustment for pensions, and restricted entitlement of working pensioners for the basic pension. At the same time, however, the payroll tax rate was also raised—further adding to non-wage labor costs.

The Federation of BiH adopted legislation that sharply cut pension and disability insurance entitlements compared to pre-war levels. The Republika Srpska has passed legislation that cuts these entitlements even further, including higher retirement ages and lower pension accrual rates. However, statutory entitlements for pension and disability insurance continue to exceed contributions in both Entities and both schemes continue to accumulate debts towards their pensioners.

First steps were also taken by Romania to address the concern that, over recent years, a rising dependency ratio had required ever higher contributions. A new public pension law will provide a limit on the replacement rate and increase the minimum retirement ages starting next year. To alleviate the effects of recent cuts in pension spending, the government “recorrelated” pensions in such a way that higher adjustments were granted to the lowest pensions. However, such a measure also weakened the contribution-benefit link and therefore may create disincentives for those earning higher wages to participate in the system.

With regard to social health insurance only limited initiatives can be observed within the group. Lithuania’s recently established Health Insurance Fund (HIF) is a welcomed step in the process of redefining the government’s role. However, the allocation of HIF resources to territorial patient funds is based on historical usage patterns, and thereby reinforces the traditional overemphasis on secondary and tertiary care. Moreover, the free provision of public health care services may be too expensive, and is likely to lead to overconsumption of services. At the same time, HIF’s financing is complex and potentially inefficient. Based on a combination of payroll taxes, income taxes, and budget transfers, it requires collection by three agencies and results in little relationship between contributions and benefits.

Slovenia has made some efforts to contain health care costs, however, management and cost control in hospitals remains a problem. As a result, Slovenia’s Health Insurance Institute is projected to move to a deficit of up to 1 percent of GDP in the coming years. Romania just established a National Health Insurance House, yet further concrete steps still have to materialize. Moldova still must deal with the problem that spas, sanatoria, and over-staffed and under-utilized hospitals absorb a large share of the Social Insurance’s resources.

Most of the countries in this group have made some progress in addressing the urgent task of reducing the number of subsidies and privileges, and enhancing the targeting of social transfers. In Moldova—where in the past there have been more than 30 categories of privileged consumers entitled to reduced tariffs of various services—the housing, transportation and utilities privileges were eliminated in 1999, and instead, money compensations will be provided directly to the most vulnerable groups. However, further steps are necessary to separate the sources and uses for social insurance and social assistance, and to eliminate privileges such as additional pension payments to former parliamentarians and government members.

After a crisis in the country’s orphanages, the Romanian government took several steps to better target its social transfers. All functions related to the care of children and orphans were consolidated into a new agency, which will receive earmarked revenue no longer fungible for other expenditure. Legal actions are also planned to consolidate the institutional responsibility for the delivery of social assistance benefits, to improve the allocations for a means-tested minimum income support scheme, and to better target child benefits to larger families.

In Lithuania the government raised electricity tariffs and plans to abolish the existing VAT reimbursements for energy consumption. These are essential measures to make the power company’s operations viable, prepare it for privatization, and set incentives for a more efficient use of energy. To mitigate the adverse effects of the increase in energy prices on the poor, targeted subsidies will be provided for low-income households.

Bosnia Herzegovina’s social welfare system needs to heighten its responsive to the variety of social risks faced by all. Although the high concentration of expenditure on military invalids and their surviving families in part reflects the large number of war victims, their entitlements have been particularly generous compared with those for civilians in similar situations (e.g., work related disability). The pension and disability insurance legislation mentioned above have been a step in the right direction, however, social assistance for those not covered by social insurance programs has received scant attention so far. Severe regional variations exist with regard to benefits and public services for this vulnerable group.

Despite its advanced status in the transition process, Slovenia has not made much progress in targeting its social transfers. Its portfolio of transfer programs comprises over 60 different benefit types which are mostly not-means tested. To stop the rapid expenditure growth in this area, the streamlining of the transfer programs and increased means-testing will be indispensable.

With regard to education, much remains to be done also. Lithuania’s public education objectives are too expansive, including free education at all levels and a day-care system that provides little “education”. The student-to-teacher ratio is low, in particular at pre-school and higher-education levels. Also, it could be argued that the mix of educational outputs does not adequately meet the educational needs of the economy. Similarly, Moldova could save expenditure by bringing student-teacher ratios to international standards, and by containing utility costs (e.g., by reducing the school week to 5 days, shifting school vacations to winter, and installing energy meters).

Bosnia Herzegovina recently has made significant strides in labor market reform. A package of radical amendments to labor legislation is designed to simplify the dismissal of excess labor, facilitate fixed-term and part-time employment, lower unemployment benefit entitlements to more affordable levels, and increase competition and flexibility in new employment.

Lithuania has also made some efforts in reducing distortions in the labor market. It increased the number of days of sick leave that is paid by the employer, thus reducing the incentive to abuse sick leave benefits. On the negative side, however, Lithuania’s Employment Fund resources still mainly go to job creation and training programs, while unemployment benefits are low and not linked to past contributions. By the same token, Slovenia’s extensive menu of active labor market policies must be reassessed with respect to its cost-effectiveness and potential negative labor market implications. Also, Slovenia’s very restrictive employment protection legislation could be liberalized in several areas.

Many reforms outstanding: Albania, Czech Republic, Georgia, Slovak Republic

As in the other groups, the background and the position in the overall transition process of this group’s countries is very diverse. The common characteristic is, however, that in comparison to the previous group of countries, their social sector reforms have not gone far. Much further efforts will be needed to address their problems in many social areas, notwithstanding that these problems vary strongly among the group’s countries.

The pension system of the Czech Republic, an advanced transition country, for example, does not face an immediate problem, but unless changes are made, an aging population will cause pension expenditure to rise at a very fast rate sometime after 2010. At the same time, the Czech pension system remains relatively generous, compared with other advanced transition economies that have undertaken comprehensive reforms (e.g., see Bulgaria and Poland). Statutory retirement ages are relatively low, penalties for early retirement remain limited, non-contribution periods are widely accepted, the minimum contribution period for a full pension is still quite short, and the time period for assessing earnings in computation of the pension base has lengthened only very slowly. Against this background, parametric reforms, such as introducing a higher and uniform retirement age for men and women, imposing actuarially fair penalties for early retirement, and bestowing less generous non-contribution periods, could go a long way.20 At present, however, none of such changes are underway.

In the past, the Slovak Republic’s pension system had also seemed free of an immediate financial threat. Similar to the Czech Republic, the effect of population aging has not yet kicked in, and pension expenditure had been in decline, absorbing a smaller share of GDP than in other central European countries. This development had been primarily attributable to a declining replacement rate, which in turn resulted from the absence of an indexation mechanism, and a contribution-benefit link that lets benefits decline progressively with higher wages.

However, generous pension adjustments made recently, combined with weak contribution collections, have put significant strains on the finances. To prevent a further worsening, reform measures need to be considered now—before the long-term demographic difficulties will further exacerbate the problem. Again, raising the currently low retirement age, imposing penalties for early retirement and tightening eligibility to disability benefits are among the possible remedies. It is noteworthy in this context that even in the past, when Slovak pensions benefits were still considered low, the number of pensioners and amount spent on early retirement and disability pensions was nevertheless very high compared with the OECD countries.

In contrast to the Czech and the Slovak system, Albania’s pension fund is facing financial imbalances already today, albeit for different reasons than in most other countries. Its financial difficulties neither result from the system’s generosity, nor are they rooted in the demographic situation. In fact, the country’s demographics rank among the most favorable in Europe, with less than one tenth of the population being age of 60 and over. Yet, Albania’s system dependency rates are among the worst, because it has the lowest ratio of active contributors relative to the working age population.21 Consequently, contribution revenue is very low. While administrative efforts to widen the participation rate in Albania’s social security system and to reduce opportunities for evasion are underway (see below), the results will probably need some time to materialize and will also hinge on the strict enforcement of the new rules.

With perhaps the exception of the Czech Republic, the countries in this group urgently need to streamline, better target, and, in part, newly build their social assistance programs. In the Slovak Republic, many of the over more than a dozen different state benefit programs overlap with the even higher number of social assistance programs administered under the local safety net. While spending on these programs has been rising, and the sum of social security contributions and income taxes is higher than in other Central European countries, the wide diversity of programs facilitates abuse and has complicated their administration. Strengthened eligibility criteria and means testing could generate substantial savings by reducing the unwarranted accumulation of benefits in many non-poor households.

In Georgia and Albania, basic social protection remains rudimentary and inadequate in many areas. In Albania, large migration movements caused urban poverty to increase, as evidenced by the spread of slum-like communities around major cities. In Georgia, unrealistic budgeting created a situation of permanent underfinancing in which the government resorted to ad hoc procedures and sequestration as a form of expenditure management. The rapid accumulation of expenditure arrears on wages, but also pensions and social allowances, has had a direct adverse impact on the incidence of poverty. Neglecting basic safety net expenditure has, in particular, hurt the urban poor, since they lack informal coping mechanisms, such as the subsistence farming available to the poor in rural areas.

In this environment, the first step has been to establish expenditure control mechanisms that set spending priorities for social safety net functions, and prevent the further accumulation of arrears on wages, pensions, and social transfers. As a second step, however, the components of social spending must be scrutinized with regard to their cost-effectiveness. For example, the various benefits provided under Georgia’s Internally Displace Persons program—in size second only to the pension program—are considered to be poorly targeted, and the electricity tariff discounts are also an inefficient way to protect poor households. In the case of Albania, the targeting of the existing cash-benefit program needs improvement, and the country needs to develop a system of community-based social care services.

Improvements in revenue performance through strengthened tax and customs administrations, again, are most pressing in Georgia and Albania, where the maintenance of minimum social spending levels is at risk due to a lack of available finances. Albania has started to give social security identification numbers to all individuals in order to increase compliance with contribution obligations. Moreover, new amendments to the social security law will strengthen the authority of inspectors in collecting contributions. In Georgia, similar measures would also be advisable. The country’s revenue is estimated to be only half of its potential. Thus better administration and an intensified collection effort could bring substantial improvements in revenue performance.

For the Czech and the Slovak Republic it must be emphasized that social sector reforms could help to reduce the existing distortions on their labor markets. First, payroll contributions for the social insurance funds are high in both countries. Combined with the personal income tax, they drive a large wedge between labor costs and take-home pay, creating significant disincentives for both labor supply and demand. All the above suggestions to contain costs by a more efficient use of available resources would help to take the pressure off tax and contribution rates, and thus would indirectly contribute to combat the high unemployment rates prevailing in these countries. Second, there are strong indications that the sharp increase and the regional pattern of unemployment in the Czech Republic, albeit initially set off by corporate restructuring, has been accentuated by generous social assistance and rent controls that discourage regional labor mobility.

Substantial reforms are also needed in the health and education sectors. The Slovak Republic has taken initial steps in recent years to control health care costs, yet the very high level of the health fund’s expenditure and revenue arrears, and its high and persistent deficits show that serious problems remain. An oversupply of physicians and hospitals, increasing pharmaceutical prices, and few constraints on the utilization of services have caused expenditure to raise rapidly. A strategy to address the financial imbalance, and to alleviate the risk of a deterioration in the quality of services, could include: introducing deductibles and higher co-payments, continuing to reduce the costs of health facilities, developing a system of risk-related premiums with transparent cross-subsidization, and reviewing the collection system and increasing compliance with contribution payments.

In Georgia, reforms should address the intra-sectoral allocation of spending on health care and education. Too much emphasis is still put on curative health care, although it is preventive health care that is more needed by the poor. In the same way, half of the executed education budget is allocated for universities, and only one fifth to primary and secondary education.

Albania’ Interim-Poverty Reduction Strategy Paper points in the right direction. Priority is given to access to primary health care, with special emphasis on maintaining high immunization rates and the strengthening of maternal, prenatal and neonatal care. Furthermore, the Paper recognizes the need to reverse the trend of declining enrollment in secondary education, while preserving the almost full enrollment in basic education, and assessing its quality and effectiveness. Yet all will depend on the execution of this strategy. Thus far, Albania’s main health indicators compare very unfavorably with regional averages, enrollment in secondary education dropped dramatically, and budgetary expenditures on both health and education are substantially lower than in neighboring countries.

C. Looking Ahead: Guidelines for Future Reform

In light of the past developments and the current reform efforts analyzed above: What general guidelines can be derived for future reforms in the various social sectors of transition countries?

Social insurance

Public pension policy must direct the thrust of reform measures towards closing the gap that has emerged between demographic dependency ratios and pension systems dependency ratios. On the one hand, this requires a reduction in the number of beneficiaries by those who retire earlier than the official retirement age (limiting early retirement options) or who are not truly disabled (tightening eligibility to disability pensions). It may also call for considering increases in the official retirement age.

On the other hand, ways must be sought to raise the share of active contributors among the working age population. To break the vicious circle of high statutory rates and flight out of the formal sector, expenditure savings through reducing benefit levels might have to be considered in addition to those measures aiming at a reduction in the number of beneficiaries. This in particular applies to countries where demographic dependency ratios are worsening rapidly.

Often replacement rates have been relatively generous and could be reduced—at least for certain groups or programs—without causing social hardship. The pension formulae commonly used by pay-as-you go schemes can be altered in several ways: the period of working years used to calculate the pensionable wage can be extended (often the formulas only use the last or best years of earnings as a base). Similarly, the period required for the minimum pension can be lengthened (after the minimum period, an accrual factor typically raises the pension for every additional year of work).

Such adjustments in the pension formula would also highlight the link between contributions and benefits, and thereby further remove incentives to evade contribution payments. For example, a relatively high pension benefit bestowed after a short period of minimum contribution years in combination with a low accrual factor, creates an incentive to no longer comply with contribution obligations once the minimum contribution requirement has been fulfilled.

In the same vein, the benefit-contribution link could be strengthened by increasing the share of the employee’s contribution. Indeed, this would neither have to decrease the workers’ take home wage nor the employers’ total labor costs, if the workers gross wage is exactly raised by the amount that the employers contribution obligation is reduced. This simple shift of a non-wage labor cost element into the workers’ wage compensation would indicate to workers more clearly who finally bears the costs of the pension system and prevent common misperceptions, for example, regarding the treatment of self-employed compared to regular employees.

Accompanied by efforts to strictly enforce collection of contribution liabilities, implementation of some, if not all, of the above measures would help to bring the working-age population back into the system as active contributors, and thus close the gap between system- and demographic dependency ratios.

With regard to unemployment schemes, measures should be taken to reduce the often still large portion of contributions spent on low-priority programs and benefit administration. Although unemployment benefits must strike a balance between social protection and disincentive effects, it should be ensured that benefits are adequate in relation to prevailing wages, and are paid in time as well as for a duration that allows for job search.

An expansion of coverage may be warranted in those countries where only a small share of the unemployed actually receive assistance in any form. However, a distinction between assistance and insurance should be drawn. Unemployment insurance schemes should only provide benefits to those who have contributed, lest the incentive structure becomes distorted. Those who have not contributed must be referred to social transfer (welfare) programs that provide means-tested unemployment benefits financed from general tax revenue.

Social transfer (welfare) programs

Social welfare programs should be clearly distinguished from social insurance schemes. In contrast to insurance schemes, the risks covered by welfare programs are of such a nature that they should be financed by general revenue. Hence, there is no individual contribution-benefit link to be strengthened. Instead, welfare programs must ensure that their benefits are actually received by those identified as genuinely needy. In other words, welfare benefits should be well-targeted.

General subsidies for food stuffs, energy, and other goods lack any targeting at all. They should therefore be abolished and replaced by cash transfers to those consumers who otherwise would not have the means to purchase sufficient amounts of the good.22 Although many of the existing transfers and privileges in transition countries are targeted towards certain groups, they are not well-targeted. First, many groups traditionally granted extensive privileges (judges, parliamentarians, internal security personnel) have become selected for their political influence, not their vulnerability to poverty. Second, other groups are too broadly defined and thus comprise not only those who are in need of support. Free public transportation or reduced tariffs for communication and electricity for all pensioners, for example, help well-off pensioners as well as poor pensioners. In general, criteria such as age, occupation, being a war veteran, or even the mere number of children do not necessarily imply destitution, and may lead to mistargeting.

Where the poor can not be easily identified as entire groups, the eligibility to transfers and allowances is best determined through means-testing of individuals or households. To the extent possible, such a means test should not only be based on official wage income, but take into account all sources of income. Otherwise, even a means test carries the risk of mistargeting, and could create disincentives for the labor supply, in particular when the informal sector is large (e.g. in case of the housing subsidy program in Ukraine). This can pose difficult implementation issues, in the absence of an adequate information base (see Box 2).

Information Requirements—a Potential Constraint for Social Policy Design

Weak information bases and the lack of adequate data systems can pose substantial problems when designing policy:

  • First, the absence of data makes it difficult to conduct the necessary empirical research serving as the basis for policy advice. Indeed, almost all of the available empirical studies note that data limitations constrain their ability to shed more light on the critical links between public spending and social development. To give an example, most non-OECD countries have not yet gathered the basic data on the inputs and outputs of their health systems (total health spending, the poor’s access to health care, the efficacy of resources used, operations in the private market etc).

  • Further, an insufficient information base may impair the implementation of policies that otherwise appear clearly desirable. Means testing in order to better target social benefits is one example: the lack of data on the true income situation of households may result in unwarranted payments, and thus undermine the very idea of means testing.

  • Finally, data weaknesses limit the scope to monitor the implementation of social policies through intermediate indicators.

Closely related to the information problem is that of administrative capacity constraints. More comprehensive and targeted interventions are often administratively burdensome, and therefore might overwhelm the existing capacities. There is no general rule, however, but the advice to carefully prove in each case whether a policy that might look good on paper is in sink with administrative realities.

Health care and education

Reforms in both the health care and the education sector have been generally slow so far, even though efficiency improvements are urgently needed given the reduced availability of financing. Studies show that the impact of health care and education systems does not necessarily depend on the level of public spending, but as much on the intra-sectoral allocation and the composition of spending.23 Further, education and health care spending in the transition economies has been high but inefficient (see Box 3). Over-staffing, duplication of facilities, and an overemphasis on non-essential services has inflated spending in both sectors. Therefore, the retrenchments of overall spending observed in these sectors often at least partly results from the shedding of redundant staff. Indeed, the non-wage expenditure often falls much less than the overall expenditure.24

Along these lines, education policies first must ensure that the available resources are adequately allocated within the sector. Tertiary education programs should not absorb excessively large shares of the education budget at the expense of primary and secondary education. Also, resources must be shifted away from the public pre-school systems which traditionally have been very large in size and scope. Second, the delivery cost of education services can be lowered in many areas, for example, through consolidation of schools (e.g. Armenia) and reductions in the excessive number of educational staff, both teaching and non-teaching. Such measures are warranted in light of high shares of non-teaching staff, and low workloads for (secondary) school teachers—reflected in the small class sizes and low student-teacher ratios in many transition countries. Also, in some regions simple changes, such as shifting school vacations to the winter, could help to contain utility costs, which weigh heavy on educational budgets. Third, the quality of spending can be improved through changes in the school curriculum, and by ensuring that textbooks and other essential teaching materials are provided.

The Government’s health care services should in general focus on primary health care services and seek to strike the right balance between preventive and curative care. Considerable resources could be set free by consolidating facilities and eliminating excessive hospital beds. Public health and disease prevention must be given priority over maintaining the jobs and protecting the incomes of government-employed health practitioners. Where necessary such rationalization measures will be accompanied by retraining programs.

Policy Advice for Healthcare and Education Spending—A Highly Difficult Issue

The policy issues related to the health care and education spending are as important as they are complex. At the outset, there is the general recognition that public spending in favor of these sectors can boost economic growth, promote equity, and reduce poverty. The enhanced HIPC—which emphasizes the link between debt relief and increased expenditure on human development – and the new PRGF—are clearly designed along this line. In fact, the World Bank and the IMF agreed that in implementing these programs, a clear set of outcome-oriented goals for social indicators should be used to promote well-designed poverty reduction strategies and to ensure an effective monitoring of their implementation.

In the case of health and education, final outcomes indicators are, for example, higher literacy rates and lower infant mortality rates, which could be monitored through intermediate indicators, such as the net primary education enrollment rate, completion of the 4th grade of primary education, immunization coverage, and the percentage of births attended by trained staff.

However, empirical work analyzing the link between aggregate public spending on education and health care and improvements in the related social indicators has yielded conflicting evidence. Gupta, Verhoeven, and Tiongson (1999), and Guin-Sui, Yamada, and Corbacho (2000) provide an overview of the results yielded by such empirical studies. In the case of education, the relationship was often weak, indeed, variables like GDP per capita and the population age distribution showed a much stronger impact. In the case of health, empirical research has been mixed. Although some authors find evidence of a positive association between overall health spending and the respective outcomes indicators, others find the link to be statistically insignificant, thus making clear cut policy advice difficult.

Fortunately, as shown by Gupta, Verhoeven, and Tiongson (1999), the evidence that the intra-sectoral composition of health care and, in particular, educational spending matters is stronger. Thus recommendations for a reallocation of resources on primary education and health care seem well warranted.

However, many caveats remain with regard to estimating the budgetary resources actually needed to achieve certain social targets. As already noted, next to government spending many other difficult to control variables play an important role in the sectors’ performance. In particularly, private sector spending on health and education must be taken into account; Could increased public spending crowd out existing private spending in the respective areas, and might the public resources be used less efficiently? Thus, while research might provide compelling reason for governments to shift their resources towards basic health care and primary education, it can not yield conclusive evidence that doing so will actually help to improve the population’s health and education status.

Finally, the ability to actually monitor improvements in these areas during a program is limited. The recent research by Guin-Sui, Yamada, and Corbacho (2000) does provide evidence for the positive effects of higher levels of social spending on social indicators, but the authors also point out that these effects may not be immediate. Indeed, the time span needed to obtain measurable results might exceed the typical duration of today’s programs for the poorest countries. Thus, possible progress achieved through altered expenditure strategies might not be recorded by the intermediate and final social indicators selected to monitor a program’s performance.

The financing of health care is probably one of the most complex social policy issues governments face.25 To replace the general-revenue financing of health services, health insurance plans have been or are being established in almost all transition economies, however, with mixed success. Such insurance plans must be carefully imbedded in an coherent health policy framework that addresses not only the problem of securing adequate financing but also the macro-organizational imperatives, the incentive structure and the regulatory framework. Thus, issues such as resource allocation methods, price setting, volume control etc., must be addressed in chorus. Without linking all the components of the health care system, potential market failures that threaten any health insurance scheme may arise. Health funds operating under fee-for-service payment systems have been one example: The fee-for service method encouraged physicians to prescribe inappropriate tests, treatments, and medications, thus, in absence of further volume controls, insurance plan’s expenditures soared.

One possible remedy would be to switch from the fee-for-service payment to per-case or capitation-based methods. The latter are particularly advisable in cases where thus far resource allocations for hospitals have been based on the number of (occupied) beds. Indeed, long hospital stays often are the result of the incentive structure set by the budget allocation mechanism rather than being motivated by medical concerns. Thus, capitation-based allocations might result in a reduction of the average length of hospital stays without impairing curative care. Similarly, adopting more effective treatment protocols, including the more rational use of prescription drugs and the use of generic drugs, can entail substantial costs savings without negatively affecting the quality of health care. Moreover, public insurance plans should principally only cover basic levels of health services to all insured, patients who demand services beyond this level can purchase private supplementary insurance.

Finally, fighting corruption is an important challenge. Gupta, Davoodi and Tiongson (2000) conclude in their recent study on the effects of corruption in the health care and education sector that improvements in indicators of health care and education services do not necessarily require higher public spending. Their results highlight that it is equally, if not more, important to institute transparent procurement procedures and enhance financial accountability of public spending. Furthermore, reduced levels of corruption in the provision of services are likely to improve their quality. Facilitating private sector entry into the provision of these public services would help to curb monopoly power of government service providers and limit their ability to charge bribes.

Tax administration

Tax administrations can contribute to social reform by strengthening compliance in the collection effort of payroll and other taxes, and by clearing arrears. In this context, it is not only necessary that existing enterprises and government agencies honor their statutory obligations, but also that new enterprises are registered and audited for enforcing taxes. The tax administrations in transition economies mostly capture only a fraction of the activity in the fast growing sectors dominated by new small private producers. Apart from making sure that tax administrations are adequately staffed, governments in many cases can also rationalize the administrative process. Unified agencies could integrate the collection, audit, and enforcement of taxes and social insurance contributions. Thus, redundancies would be avoided and the entire revenue system is more easily overseen. There is little reason for social insurance funds to maintain their own collection capacities, parallel to the general tax collection agency.

Labor market

Labor market policies are closely interrelated with social policies and anti-poverty strategies. In contrast to many developing countries, poverty in transition economies is predominantly transient in nature. In other words, the vulnerability to poverty is high because the volatility of household consumption is high, which, in turn, mainly results from a lack of stable employment. The high unemployment prevailing in many of the transition economies makes it difficult for households to improve their living standard and thus escape from poverty. Also important in this connection, the research work on social spending highlights that for targeted spending to have a considerable payoff, the benefits from improved basic social services have to be accompanied by income-earning opportunities.26

The existing labor market regimes in many of the transition economies appear to be obstacles to job creation. In general, labor laws need to strike the right balance between protecting workers’ rights on the one hand and allowing for sufficient labor market flexibility on the other. The existing labor laws in many transition economies, however, provide for very generous vacation and maternity leave benefits, extensive sick leave, and job protection. This makes the process of hiring and terminating workers cumbersome and expensive. In an effort to increase labor market flexibility and make labor codes more appropriate for the prevailing labor market conditions, countries may have to liberalize overly restrictive employment protection legislation, rethink minimum wage practices, and allow for flexible fixed-term contracts and other tools conducive to job creation.

Facing high structural unemployment, some countries have pursued active labor market policies with notable success. Over time, they have developed an extensive menu of instruments, including on- and off-the job (re)training, job subsidies, and public works. Nevertheless, such policies should be continuously reassessed in view of their cost and the their effectiveness. In particularly public works programs can have important implications for the labor market. For example, if the wage for program participants does not remain below the prevailing market wage for unskilled labor, public works programs may actually attract already employed workers.


  • Andrews, Emily S., Rashid, Mansoora, 1996, The Financing of Pension Systems in Central and Eastern Europe, World Bank Technical Paper No, 339.

    • Search Google Scholar
    • Export Citation
  • Branco, Marta de Castello, 2000, Pension Reform in the Baltics, Russia, and other Countries of the Former Soviet Union (BRO), IMF Working Paper WP/98/11.

    • Search Google Scholar
    • Export Citation
  • Cangiano, Marco, Cottarelli, Carlo, Cubeddu, Luis, 1998, Pension Developments and Reforms in Transition Economies, IMF Working Paper WP/98/151.

    • Search Google Scholar
    • Export Citation
  • Fischer, Stanley, Sahay, Ratna, 2000, The Transition Economies After Ten Years, Working Paper WP/00/30.

  • Guin-Su, Maria Teresa, Yamada, Gustavo, Corbacho, Ana, 2000, The Effects of Public Spending in Education and Health on Social Development Indicators: A Panel Data Analysis, IMF Working Paper.

    • Search Google Scholar
    • Export Citation
  • Gupta, Sanjeev, 1998, Economic Transition and Social Protection—Issues and Agenda for reform, IMF Paper on Policy Analysis and Assessment.

    • Search Google Scholar
    • Export Citation
  • Gupta, Sanjeev, Davoodi, Hamid, Tiongson, Erwin, 2000, Corruption and the Provision of Health Care and Education Services, IMF Working Paper WP/00/116.

    • Search Google Scholar
    • Export Citation
  • Gupta, Sanjeev, Verhoeven, Marijn, Tiongson, Erwin, 1999, Does Higher Government Spending Buy Better Results in Education and Health Care?, IMF Working Paper WP/99/21.

    • Search Google Scholar
    • Export Citation
  • Gupta, Sanjeev, Christian Schiller, Henry Ma, 1999, Privatization, Social Impact, and Social Safety Nets, IMF Working Paper WP/99/68.

  • Gupta, Sanjeev, Dicks-Mireaux, Louis, Khemani, Ritha, McDonald, Calvin, and Verhoeven, Marijn, 2000, Social Issues in IMF-Supported Programs, IMF Occasional Paper No. 191.

    • Search Google Scholar
    • Export Citation
  • Hsiao, William, 2000, What Should Macroeconomists Know About Health Care Policy—A Primer, IMF Working Paper WP/00/136.

  • IMF, 1998, Should Equity be a Goal of Economic Policy?, IMF Fiscal Affairs Department, Economic Issues Series, 16.

  • Keane, Michael, Prasad, Eswar, 2000, Inequality, Transfers and Growth - New Evidence from the Economic Transition in Poland, Working Paper WP/00/117.

    • Search Google Scholar
    • Export Citation
  • Kolodko, Grzegorz, 2000, Globalization and Catching-Up-From Recession to Growth in Transition Economies, IMF Working Paper WP/00/100.

  • Laursen, Thomas, 2000, Pension System Viability and Reform Alternatives in the Czech Republic, IMF Working Paper WP/00/16.

  • Milanovic, Branco, 1998, “Income, Inequality, and Poverty During the Transition from Planned to Market Economy,” World Bank Regional Sector Study 1998-2.

    • Search Google Scholar
    • Export Citation
  • Rutkowski, Michal, 1997, Security Through Diversity—Reform of the Pension System in Poland, World Bank.

  • Sala-i-Martin, Xavier, 2000, Transfers, Social Safety Nets, and Economic Growth, IMF Working Paper WP/96/40.

  • Schneider, Friedrich, Enste, Dominik, 2000, Shadow Economies Around the World—Size, Causes, and Consequences, IMF Working Paper WP/00/26.

    • Search Google Scholar
    • Export Citation
  • Tanzi, Vito, Tsibouris, George C., 2000, Fiscal Reform Over Ten Years of Transition IMF Working Paper WP/00/113.

  • World Bank, 2000, Kyrgyz Republic—Review of Social Expenditures, (July), in two volumes.


Paper presented at the East-West Conference on “Completing Transition: The Main Challenges,” November 6–7, 2000, in Vienna, Austria, sponsored by the Oesterreichische Nationalbank and the Joint Vienna Institute.


While there is consensus that an extreme inequality of income, wealth, and opportunity is undesirable and that efforts should be made to improve the situation of the poorest members in society, the question of what exactly constitutes a “fair” distribution of income and whether greater income equality is desirable for its own sake must remain open. IMF (1998).


For the evolution of IMF’s social policy advice, see Gupta et al. (2000).


For the main economic and social developments during transition see: Fischer, Sahay (2000), Kolodko (2000), Milanovic (1998), Tanzi and Tsibouris (2000).


For the dynamics of pension benefits see Andrews and Rashid (1996), Rutkowski (1997), Cangiano, Cotarelli and Cubeddu (1998), Castello Branco (1998), Keane and Prasad (2000).


The average public spending on education and health care relative to GDP in the BRO countries fell from 6.4 and 3.5 percent in 1992 to 4.1 and 2.7 percent 1996, respectively. Yet, even these figures understate the real resource squeeze in these sectors in light of the large contraction in real GDP since the onset of the transition process. Gupta (1998).


For a discussion of the social impact of privatization see Gupta, Schiller and Ma (1999).


In 1993, on average 60 percent of those in CEE countries who received pension benefits for their first year had not yet reached the age of 60. In Croatia, an extreme case, 90 percent of the first-year pensioners were younger than 60. Similarly, in most CEE countries between 20 and 25 percent of all pensioners receive disability benefits, led by Poland where, in 1993, nearly 37 percent of all pensioners received disability benefits. See Andrews and Rashid (1996).


It has been estimated that in the first half of the 1990s the share of the unofficial economy in CEE countries increased on the average from less than 18 to over 31 percent of GDP. In BRO countries this development was even more pronounced, where the average share is estimated to have grown from less than 17 to over 43 percent of GDP. For an overview of estimates see Schneider and Enste (2000).


As calculated for the BRO countries in Castello Branco (1998).


In principle, it is not uncommon for revenue systems to have certain institutions charge fees from those who use specific services and therefore should cover the related costs (equivalence principle). If well designed, including provisions for those who may not be able to pay the fee, a user-fees system has incentive advantages over pure tax-financing.


The sample of countries comprises: Albania, Bosnia Herzegovina, Bulgaria, Czech Republic, Georgia, Lithuania, Moldova, Poland, Romania and Slovak Republic, Slovenia. Information sources are mainly internal IMF staff reports and IMF Working Papers.


Countries outside the selected sample who are undertaking comprehensive pension reforms are, for example, Estonia, Latvia, and Kazakhstan.


The Social Fund’s “netting out” procedures have at times taken bizarre forms indeed: Pension arrears not only canceled pensioners’ overdue electricity bills, which in turn canceled the electricity company’s contribution arrears, but payroll taxes were sometimes also settled in goods like umbrellas, which were then given to pensioners in lieu of their benefits.


For simulations see Laursen (2000).


For example, in 1993 the Kyrgyz Republic replaced the general consumer subsidy for bread with cash transfers to pensioners and families with many young children.


Gupta, Verhoeven, Tiongson (1999), Guin-Su, Yamada, Corbacho (1999).


See, for example, World Bank (July, 2000): Kyrgyz Republic – Review of Social Expenditure.


For the following see Hsiao (2000).