12 Poland: The Social Impact of Transition

Ke-young Chu, and Sanjeev Gupta
Published Date:
April 1998
  • ShareShare
Show Summary Details
Gerd Schwartz

Poland was one of the first economies to undertake the transition from plan to market. This chapter looks at the safety net that existed at the outset, describes how it was reformed during the transition, and analyzes how the poor and needy fared during this process.

Social Safety Net at Outset of Transition

Poland’s pretransition social protection arrangements centered around an extensive social insurance system, which, inter alia, provided for old-age, disability, and survivors pensions; health care; work injury benefits; sick pay; family allowances; and maternity benefits. In addition, there existed a limited system of local social assistance to support the poor. While, on the surface, this does not look unlike the arrangements found in many market economies, the differences were in fact striking.

A basic feature of Poland’s pretransition social protection arrangements stems from the central role of labor in socialist societies; participation in the labor market was viewed as almost obligatory, also because access to many social benefits depended on being in the labor force. While employment and social protection policies aimed at providing a job guarantee to every citizen, the distinction between wages and cash benefits was blurred, with wages largely set to achieve distributional objectives. There was considerable hidden unemployment that disguised the need for unemployment benefits, job counseling, retraining, or social assistance.

A basic feature of Poland’s pretransition social protection arrangements stems from the fact that poverty was essentially viewed as a distinguishing feature of capitalism. Explicit basic income and emergency support policies were made largely superfluous by a combination of fixed prices, extensive subsidies for a large number of basic consumer goods, and substantial cash transfers in the form of family allowances and other benefits. In addition, many benefits were administered and provided directly by state-owned enterprises (SOEs). These included free vacations at designated resorts, housing, child care, access to consumer goods, and other in-kind benefits.

Who Were the Poor at the Outset of Transition?

Poland’s income distribution before the outset of transition was similar to other socialist economies. It can be characterized by seven stylized facts (Milanovic, 1992a): (1) income inequalities were less pronounced than in market economies even though there existed problems of differentia] access to benefits; (2) rural incomes exceeded urban incomes (in Poland by about 16 percent); (3) property income was insignificant; (4) in-kind benefits were more important than in market economies while wage income was less important; (5) self-employment income, including in-kind consumption, was relatively more important than in many Western economies, reflecting a large agricultural sector; (6) similar to Western economies, cash transfers accounted for about one-fifth of gross income, but, unlike in Western economies, these cash transfers were distributed almost equally on a per capita basis; and (7) direct taxes played almost no role in redistribution.

The various existing studies on poverty in pretransition Poland all tackle the standard problems of defining a poverty line, measuring the number of people that fall below this poverty line, assessing by how much they fall below the line, and estimating how long they remain in poverty. Still, measuring the extent of poverty in pretransition Poland is complicated by the fact that the definition of living standards is somewhat abstract in economies with chronic shortages of goods and services, as having a higher income does not necessarily imply having a higher living standard (Graham, 1993). While there is no doubt that income in pretransition Poland was much more equally distributed than in market economies, the available data do not include various privileges enjoyed primarily by the nomenklatura and enterprise-specific in-kind benefits, such as subsidized vacations at company facilities, food, housing, and better access to scarce or subsidized goods.

Given these important limitations, the available evidence shows a rather consistent pattern of the degree to which different population groups were affected by poverty in pretransition Poland.1 This is not surprising as all studies use data from the annual Household Budget Survey (HBS). Hence, differences in the results reflect differences in defining poverty lines and equivalence scales for various household types, and methods of calculation. The studies by Milanovic (1992c and 1993) and Panek and Szulc (1991) are typical in this respect. Their research suggests that, during 1988–89, the poverty rate was highest among pensioners and lowest among mixed farmer and worker households (Table 12.1).2 All studies also agree that the probability of being poor was higher for single-parent households than for two-parent households, increased with the number of children, and decreased with the level of educational attainment. Contrary to some other studies, Milanovic (1992b) finds that in the 1980s there was a strong increase in the number of poor in urban areas: urban poverty increased from 7.8 percent during 1978–79 to 21.5 percent during 1987–88, largely as a result of a decline in real wages in the industrial sector during the 1980s.3 At the same time, rural poverty remained rather constant, amounting to 13.3 percent in 1978–79 and 13.7 percent in 1987–88, probably due to favorable terms of trade in the agricultural sector. All studies show a slight increase in the overall extent of poverty even before the onset of the transition.

Table 12.1.Poland: Poverty Headcount Indices by Source of Household Income
Milanovic (1992c and 1993)
All households15.217.331.534.3
Mixed farmers and workers8.07.916.121.2
Panek and Szulc (1991)
All households15.316.7
Mixed farmers and workers6.16.6
Rural population20.423.3
Nonrural population9.79.7
UNICEF (1993)
All households21.840.339.341.4
Urban households19.837.033.033.6
Rural households24.443.148.452.4
Ochocki (1993)
a.Total “objective” poverty19.424.838.7
b.Total “subjective” poverty32.635.5
Rural population38.547.4
Nonrural households129.330.0
One-person households69.263.4
Three-person households18.728.6
Six- and more person households27.936.9
“Subjective” poverty, 1990–9111.6
Mixed farmers and workers5.1
Rural households13.4
Nonrural households112.4
Sources: Cited studies; and IMF staff estimates.

Unweighted average of various nonrural categories.

Sources: Cited studies; and IMF staff estimates.

Unweighted average of various nonrural categories.

Social Protection During Transition

With the onset of transition, existing social protection arrangements quickly became unsustainable and needed to be reoriented in several aspects: to lend support to the transformation; to be aligned with the requirements of a market-based environment, and to prepare for changes in the demographic structure.

As a first step, several elements that were central to the old system of social protection, such as job guarantees, fixed prices, generalized subsidies, and various in-kind benefits, needed either to be curtailed or abolished. The extensive system of food subsidies, for example, had already de facto been abolished in 1989.

During the transition, the main elements of Poland’s system of social protection were unemployment benefits, pensions, social assistance, family allowances, sickness benefits, and, more generally, health care. At the same time as direct consumer subsidies were dramatically reduced, there was a strong expansion of the system of cash benefits. Within the cash benefit system, the brunt of the transformation-induced increases in social expenditures was borne by social insurance arrangements (mainly pensions and unemployment compensation) rather than social assistance schemes targeted to the poor or other more temporary arrangements. This was largely due to the ease of access to social security and a more attractive benefit structure. Unemployment benefits and pensions can be used as examples of the many recent changes and current problems of the system of cash benefits.4

Unemployment Benefits

As regards unemployment benefits, the Employment Law of December 1989 fully recognized unemployment and stipulated that unemployment compensation was no longer a discretionary benefit. The initial provisions were fairly generous: in general, the level of benefits was linked to the claimant’s last wage, and there was no duration limit. This invited abuse, particularly in larger cities where administration is more difficult and alternative employment opportunities are easier to come by. Often, people who were about to be laid off were given substantial wage increases to enable them to draw higher unemployment benefits. Unemployment benefits initially had the character of general income support rather than targeted assistance, particularly since many beneficiaries continued to work in the shadow economy. Also, being registered as unemployed was a way to retain access to various social insurance benefits, particularly free health care, and to continue to accumulate pension rights (Ksiezopolski, 1991). Hence, even people who intended to withdraw from the labor market generally decided to remain registered as unemployed.

Unemployment benefits underwent several reforms during 1990–93. Most important, a general 12-month duration limit was introduced in December 1991, and a generally flat-rate benefit structure at 36 percent of the average wage that prevailed in the economy during the previous quarter was introduced in February 1992. The level of unemployment benefits appears roughly in line with minimum subsistence, but more so for smaller households than for larger ones. Still, problems remain. The system does not provide an explicit minimum subsistence guarantee, and there are concerns regarding the adequacy of the indexation mechanism. Also, to the extent that the duration limit is binding, much of the burden of caring for the long-term unemployed and those withdrawing from the labor market subsequently falls upon basic income support and emergency assistance schemes, which implies a major devolution of responsibilities from the national level to the local authorities. Finally, the value of active labor market measures by the government would tend to be reduced to the extent that labor mobility remains constrained by the absence of a housing market and the existence of a general housing shortage.

The Pension System

As regards the pension system, there has been an impressive number of changes during 1990–93, but fundamental reform has yet to take place. The major changes tried to address the six main problems of Poland’s pension system:5

  • liberal eligibility criteria for early retirement and disability pensions and generous stipulations regarding the right to receive pensions while continuing to be employed;
  • a high average replacement rate, largely a result of the formula for calculating the pension base;
  • inadequate mechanisms for cost of living adjustments (COLAs);
  • persistent structural abnormalities, including occupation-specific privileges and incorporation of noninsurable risks into the social security system;
  • insufficient pension system financing, also due to evasion and the buildup of arrears to the social security system; and
  • weak administration, resulting from inefficient management, insufficient modern equipment, and inadequate regulations.

In general, it was easier to address the obvious excesses that invited abuse than to address fundamental issues.

As a result, a main unresolved problem of the Polish pension system is the reform of eligibility criteria for early retirement and disability pensions. During 1990–93 the number of pensioners increased rapidly while the number of contributors dwindled. During the four-year period from December 1989 to December 1993, the total number of pensioners grew by 28 percent, from 6.9 million to 8.8 million (Figure 12.1). Much of this increase already occurred in the early stages of the transition: between December 1989 and December 1991 alone, the total number of pensioners increased by 21 percent to 8.4 million. There was a particularly pronounced increase in the number of old-age pensioners, including early retirement pensioners, which, for the main pension fund (FUS), increased by over 36 percent during the December 1989–December 1993 period. At the same time, the number of contributors dropped from 14.8 million in December 1989 to 12.5 million in December 1993. Excluding the pension scheme for farmers, which traditionally has covered over 90 percent of its expenditures by transfers from the state budget, the ratio of contributors per pensioner (dependency ratio) dropped from 2.7 at the end of 1989 to below 1.9 at the end of 1993 (Figure 12.2).

Figure 12.1Main Pension Fund (FUS): Contributors and Pensions

(In thousands)

Figure 12.2Main Pension Fund (FUS): Dependency Ratios1

Sources: Polish authorities; and IMF staff estimates.

1 Number of labor force participants to number of retirees.

The drop in the dependency ratio has been accompanied by a drop in the average pension age, largely due to a sharp increase in early retirement. While in 1990 the average retirement age for FUS pensioners was 57 for women and 58 for men (Hambor, 1992), the overall average was estimated to have dropped to below 55 years by mid-1993. The pronounced shift toward early retirement can be attributed to four main factors: (1) early retirement, which is widely perceived as an alternative to unemployment and offers more attractive benefits than unemployment; (2) liberal eligibility criteria that, initially, did not penalize pensioners who continue to work; (3) limited supervision and enforcement capacities that make it difficult to keep track of working pensioners; and (4) anticipation of restrictive pension reforms that would adversely affect those who were to retire later.

However, it is not only the steep increase in the number of old-age pensioners that created problems for the Polish pension system but also the high level of disability pensioners. In December 1989, there were only 1.1 old-age pensioners per disability pensioner, implying that almost 50 percent of all labor market participants who had left the labor force had done so for health-related reasons before reaching regular retirement age. To some extent, the large number of disability pensioners is a legacy of socialism. Liberal eligibility criteria for disability pensions reduced some of the pressures of having to provide a job guarantee for every citizen, and thereby alleviated the extent of excessive employment. Accordingly, disability was basically defined in terms of “damage to health” rather than “inability to work.”

A second unresolved problem concerns the statutory replacement rate, that is, the average ratio of an individual pension to the work income it replaces. In Poland, the rate is relatively high, particularly considering the low dependency ratio. Related to this is the issue of cost-of-living adjustments. Until 1990, all COLAs occurred ad hoc and were constrained by the need to prevent a financial collapse of the pension system. Frequently, delays in indexation were used to generate actual replacement rates that were significantly below statutory levels. While there is no optimal level for statutory replacement rates, the arbitrary way in which the system was maintained would appear undesirable. In general, two main issues arise in connection with COLAs: the selection of the indexation mechanism, and the selection of the index itself. In 1993, pensions were indexed to average wages in the previous quarter, and the indexation mechanism was invoked when average wages in the previous quarter increased by at least 10 percent over the average wage that was used for the last COLA. While the choice of the index that was in effect in 1993 allowed pensioners to share in general productivity increases (at least to the extent that these are reflected in wages), the mechanism was somewhat tardy, and only limited, but did not prevent temporary erosions of the real value of pensions. In particular, the minimum pension, that is, 35 percent of the average wage in the economy that was used in the last COLA, did not provide an effective minimum subsistence guarantee.

Other Social Expenditures

Various other changes in social expenditures during 1990–93 largely reflected the need to contain government expenditures and were interim solutions that eventually needed further reform. Two examples may suffice. One example is the freezing in early 1992 of family allowances at a level of Zl 167,000 per beneficiary, a measure that helped to control expenditures without addressing the actual problem of finding ways to improve the targeting of these allowances. Another example is the free distribution of drugs to pensioners and health service staff that was discontinued in late 1991 (UNICEF, 1993). Whereas improving the cost-effectiveness of the health care system is important, improving the targeting of benefits would seem to be a more desirable way of containing costs than blanket reductions or withdrawals of benefits.

The Poor During Transition

What the Data Show

How did the transition affect the character and extent of poverty during 1990–93? Two main factors seem particularly relevant. First, the significant decline in output, particularly during 1990–91, affected the average level of well-being; second, new opportunities that were brought about by the transition-induced changes in the income distribution increased relative income disparities.

While it is difficult to measure the exact quantitative extent of these effects, there is general agreement on two things: first, the extent of poverty increased significantly during 1989–93; second, much of the observed increase in poverty is transition-induced. Also, there is fairly broad agreement that the increase in poverty cannot be attributed to a dramatic worsening of the income distribution, or to a contraction of social transfers, or to a retreat of the state (UNICEF, 1993). In general, the extensive safety nets in place prior to the transition have been maintained, and even extended to include unemployment compensation and social assistance benefits, which is not to say, however, that these mechanisms have been effective or efficient. Again, the general agreement on central aspects of poverty during the transition is not surprising because much of the available quantitative evidence is based on the same data source, the annual HBS.

Milanovic (1992c and 1993) finds that poverty increased significantly in 1990, the first year of the transition, from 17 percent of the population (6.6 million) in 1989 to 31.5 percent (11.9 million) in 1990 (Table 12.1). The increase was most pronounced among wage earners (workers), particularly those employed in the state sector. The extent of poverty among pensioners increased only slightly during the first year of transition but still continued to exceed any other group. These changes in poverty are mirrored by changes in income: with overall real per capita income dropping by 31 percent in 1990, farmers experienced a drop of 40 percent, and pensioners a drop of 19 percent. While the drop in the income of farmers was steep, poverty among farmers in 1990 remained below the levels found among workers and pensioners. Children appear to have been particularly severely affected during the first year of the transition: over 50 percent of all children are estimated to have lived in poverty in 1990, whereas in 1989 it was, at most, 28 percent (UNICEF, 1993).

Moreover, the poverty gap, that is, the total income needed to bring all poor households to the estimated poverty threshold level, increased by 40 percent in real terms, even though it declined by 23 percent in per capita terms (Milanovic, 1993). This may suggest that many people are just below the poverty threshold, and that estimates on the extent of poverty are quite sensitive to the definition of the poverty line. If this is correct, it would help to explain the significantly different estimates of the extent of poverty: in 1990, for example, these ranged from 19 percent of the population (Ochocki, 1993) to over 40 percent (UNICEF, 1993). Regardless of the exact extent of poverty, the principal gainers in the first year of transition were private sector entrepreneurs, some workers who shifted from the public to the private sector, and possibly property owners (Milanovic, 1992b).

During 1991–92, the extent of poverty largely stabilized at the increased 1990 level (Table 12.1). Still, the pattern of poverty underwent a marked change: while in 1990 the increase in poverty was most severe among urban households, particularly state-sector workers, in 1991 the social costs began to shift more strongly toward farmers and mixed farmer and worker households (Milanovic, 1993). Accordingly, the incidence of poverty was most severe among households that were (1) headed by pensioners, single parents, young workers, or persons with a low level of educational attainment, (2) large, with five or more members, and (3) located in small towns (below 20,000 inhabitants). While some other studies, such as UNICEF (1993), show that the transition affected urban and rural poverty to a similar extent during 1990, it is generally agreed that urban poverty did not increase during 1991–92 as rural poverty continued to increase (Table 12.1). Another interesting result is that poverty rates among pensioners remained fairly stable (albeit at a relatively high level), possibly indicating that the pension system was used as a social safety net device during the transition. Also, during 1991–93, the incidence of poverty became particularly prevalent among the unemployed: poverty among the unemployed increased to a level that was almost 50 percent higher than among the overall population.

Are the Data Correct?

Did the welfare of the population really decrease during the transition? There are at least five main problems with the available evidence. First, the discussion generally ignores the counterfactual of the transformation: at least in the long run, if not already in the short and medium term, the extent of poverty would have been likely to be even more severe in the absence of the reforms that were undertaken.

Second, the annual HBS data may not always provide an adequate reflection of reality. For one, the HBS has on occasion generated dramatic changes in poverty even before the onset of transition. For example, Panek and Szulc (1991) estimated an increase in poverty from 14 percent to 30 percent of the population during 1982–84, which is not unlike the increase from 17 percent to 34 percent of the population that was estimated by Milanovic (1993) during 1989–91. While 1982–84 were also years of crisis, and at least some increase in the extent of poverty may have been expected, the reduction in poverty to 15 percent of the population during 1984–88, as estimated by Panek and Szulc (1991), is quite surprising. These strong variations in the extent of poverty cast doubt on the reliability of the HBS data and reveal the weakness of poverty-line-based head-count analysis in this context.

Also, as already mentioned, until 1991 the HBS did not cover the self-employed, workers in the private sector, and the military and police—groups that are generally considered to be among the better off. Excluding these groups would tend to overestimate the extent of poverty and underestimate changes in the income distribution, which may help to explain the surprising stability of the Gini coefficients shown in Table 12.2. In addition, some recent estimates have suggested that about 20 percent of economic activity goes unreported. Hence, it would not seem unreasonable to assume that an equal share of income is unreported. In fact, for the fourth quarter of 1989, the Central Statistical Office (GUS) estimated that incomes reported for the annual HBS were 21 percent below the incomes reported by enterprises (Gorecki and others, 1992). If anything, this gap can be expected to have increased during 1990–93, particularly considering the growth in private sector employment. In this context, it is interesting to ask why the available evidence suggests that pensioners suffered least during the transition. A possible answer would be that pensioners are probably least able to underreport income as their main source of income comes directly from the state.

Table 12.2.Poland: General Indicators of Population Well-Being
(Average per capita consumption)
Calorie intake (calories a day)2,8912,7672,744
Milk consumption (liters a year)125.8121.3119.0114.4
Meat and fish consumption (kilograms a year)64.368.972.169.5
Bread and cereal consumption (kilograms a year)120.5118.4121.1120.6
(In percent)
Total consumption spent on food49.251.845.843.5
Population receiving social assistance
Receiving assistance regularly0.
Receiving assistance occasionally2.
Rate of children with low birth weight7.
Infant mortality rate1.
Under age 5 mortality rate1.
Preprimary education enrollment rate48.747.143.942.6
(Other as indicated)
Life expectancy at birth (years)
Income inequality (Gini coefficient)24.919.123.224.0
Sources: UNICEF (1993); and IMF staff estimates.
Sources: UNICEF (1993); and IMF staff estimates.

Third, the high-inflation environment that existed at the outset of transition is likely to have created severe measurement distortions. Average consumer price inflation amounted to about 250 percent in 1989 and 585 percent in 1990. Given the extent of price liberalization during 1989–90, the high inflation rates largely reflected step-adjustments and significantly exceeded the underlying rate of inflation. These inflation-induced distortions may help to explain some of the strong intertemporal variations in the extent of poverty: Panek and Szulc (1991), for example, estimated a drop in poverty from 23 percent of the population in the second quarter of 1989 to 11 percent in the third quarter.

Fourth, it has sometimes been suggested (Lipton and Sachs, 1990) that Polish pretransition statistics had a bias toward presenting a rosier picture of the reality, and that once this bias is eliminated, as happened during the transition, the new situation necessarily appears worse.

Fifth, even if it could be assumed that HBS data provide an adequate reflection of reality, estimates of poverty are entirely determined by the definition of equivalence scales and the poverty line for different population groups. This is a rather subjective science, as shown by the range of poverty estimates for Poland, where, for example, depending on the study, either 19 percent or 40 percent of the population lived in poverty in 1990.

Future Issues for Discussion

Given these data limitations, it would be desirable to supplement the HBS-based research. This may be accomplished in at least three ways. First, one could carry out more detailed analyses with the existing data. This could include measuring the duration of poverty and estimating the size of interhousehold transfers (or private-family safety nets). Second, explicitly subjective estimates could be employed, that is, asking household members about the income level necessary to maintain the household at its subjectively perceived poverty line, and then comparing the actual income with the subjective poverty line. Third, one could use nonmonetary indicators, particularly indicators of deprivation (e.g., nutrition, calorie intake, food consumption, and so forth), but also information concerning dwelling conditions, possession of household durables, and others.

Research along these lines has just begun. As regards the first option, recent research has shown that the duration of poverty may actually not be long for many of the poor. Ochocki (1993) has estimated, based on the Survey of Living Conditions of the Population (SLP) carried out periodically by the Central Statistical Office since 1984, that only 12 percent of the population considered themselves poor both in 1990 and 1991, although about twice as many people were considered poor in either 1990 or 1991 (Table 12.1). Still, the available evidence is not clear: from among households surveyed in both years, close to 60 percent of those who were poor in 1990 remained so in 1991, whereas evidence for 1991–92 suggests that this increased to about 80 percent, indicating that poverty shows a tendency toward becoming persistent.

Similarly, recent research has shown that private interhousehold transfers increase in response to household earning losses, and that family networks can complement governments as a means of income redistribution. Simulation exercises by Cox, Jimenez, and Okrasa (1993) on the basis of the 1986 HBS indicate that these private-family safety nets are an important part of Polish income, filling about one-sixth of the income gap left by lost earnings, and that these are well targeted to low-income households. The response was highest for households with one income earner (replacing 30 percent of the income lost) and lowest for households with more than one income earner (replacing 5.5 percent of the income lost). Thus, in many cases, increases in private interhousehold transfers could have substantially narrowed or even bridged the poverty gap.

As regards the second option, the use of subjective poverty estimates has yielded results that are not unlike the headcount indices derived from the annual HBS. Ochocki (1993) has estimated, again based on the SLP, that about one-third of all households could be considered “subjectively” poor in 1990 and 1992 (Table 12.1). These subjective poverty estimates also suggest that some of the recent research claiming that over 40 percent of the population lived in poverty during 1990–92 may overestimate the extent of poverty. Again, a high estimate of the number of poor people is the result of using a relatively high poverty line, such as the Polish “social minimum,” which is generally agreed to exceed minimum subsistence significantly (World Bank, 1993; and Graham, 1993).

As regards the third option, there are a number of largely nonmonetary indicators available. Indicators of deprivation are of particular interest for estimating the extent of poverty. In general, the available indicators do not suggest that basic needs have been threatened by the transition (Table 12.2). Probably most interestingly, the sharp curtailing of transfers to households (excluding pensions) from 8.4 percent of GDP in 1989 to 3.9 percent in 1990, and particularly the discontinuation of food subsidies, which, before the transition, accounted for nearly 30 percent of the total value of food consumption (Graham, 1993), appear to have had little effect on the average nutritional status. While milk consumption, which before the transition had a producer price that was six times above the consumer price, dropped by 9 percent during 1989–92, meat consumption, which was rationed before the transition, increased by 8 percent during the same time period. Although the energy content of the average Polish diet fell by 2 percent during 1989–91, total calorie intake remains more than adequate. It is also interesting to note that, while maintaining their calorie intake, people spent significantly less of their total consumption on food in 1992 than in 1989 (Table 12.2).

A number of other indicators have sometimes been used to estimate the extent of poverty in Poland. Usually, these indicators are not very comprehensive but may provide useful supportive evidence. For example, between 1990 and 1992, 18 percent of all kindergartens were closed, which would tend to support the argument that children are particularly affected by the transition. However, under socialism, having an abundance of child care facilities was an important element for achieving the dual policy goal of high labor force participation with full employment, and the current level of child care facilities is probably more commensurate with the requirements of a market economy.

Often, though, these alternative or complementary indicators do not achieve what they set out to demonstrate. For example, it has been argued that the fact that “at the end of 1992 rent was not being paid regularly for 43 percent of the housing units owned by cooperatives, municipalities, and employers, and that payment delays were of three months or more in about 11 percent of the cases” reflects inability to pay, which in turn is a sign of growing poverty (UNICEF, 1993). However, nonpayment is not automatically equal to inability to pay, particularly since the absence of a credible threat of eviction creates a strong element of moral hazard. In addition, while much of the housing stock is eventually to be privatized, property rights remain unclear, a situation that may also have contributed to an erosion of payment discipline.


Poland’s social protection mechanisms became fiscally unsustainable once transition got under way. The authorities’ efforts to reform the social protection system coincided with a strong increase in poverty rates in the initial stages of reform. Yet, the available data provide conflicting evidence on the adverse impact of the transition on the population’s well-being, which has complicated policymaking. Nevertheless, the Polish authorities generally succeeded in replacing the old social protection system, which relied heavily on job guarantees, generalized subsidies, and in-kind benefits with a modern system of social protection.

Note: Reprinted from Poland: The Path to a Market Economy, IMF Occasional Paper No. 113 (Washington: International Monetary Fund, October 1994), pp. 80–88.


Also see Atkinson and Micklewright (1992) for a detailed comparison of the different studies.


Until 1991, the Polish HBS sample contained four major occupational groups: households of manual and nonmanual workers employed in SOEs or cooperatives (workers), households of farmers, households with farmers and workers, households of pensioners. Households of the self-employed, private sector employees, military, and police were excluded from the sample. This was an omission that became increasingly serious as the private sector expanded. However, since 1992 all groups are represented. See Gorecki and others (1992) and Kordos (undated) for further details.


Since the basic data are the same, this finding is essentially due to definitional differences: while Milanovic defines “rural” as farmers plus mixed farmer and worker households and “urban” as workers plus pensioners, other studies use more detailed data on town size.


See Maret and Schwartz (1993) or the World Bank (1993) for detailed reviews.


See Maret and Schwartz (1994), and Schwartz (1994) for detailed reviews of pension system reform in Poland.


    AtkinsonAnthony B. and JohnMicklewright1992Economic Transformation in Eastern Europe and the Distribution of Income (Cambridge, England: Cambridge University Press).

    • Search Google Scholar
    • Export Citation

    CoxDonaldEmmanuelJimenez and WlodekOkrasa1993“Family Safety Nets During Economic Transition: A Case Study of Poland,” paper presented at a conference on Privatization and Socioeconomic Policy in Poland(unpublished; Krakow).

    GoreckiBrunon and others1992“Country Paper—Poland 1992” (unpublished).

    GrahamCarol1993“The Political Economy of Safety Nets During Market Transitions: The Case of Poland,”Research Paper Series No. 3 Research Project on Income Distribution During the Transition (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    HamborJohn1992“Issues in Eastern European Social Security Reform,”Research Paper No. 9201 (Washington: U.S. Treasury Department).

    • Search Google Scholar
    • Export Citation

    KordosJan“Poverty Measurement in Poland” (unpublished; Warsaw: Central Statistical Office).

    KsiezopolskiMiroslaw1991“Labour Market in Transition and the Growth of Poverty in Poland,”Labour and SocietyVol. 16No. 2 pp. 17592.

    • Search Google Scholar
    • Export Citation

    LiptonDavid and JeffreySachs1990“Creating a Market Economy: The Case of Poland,”Brooking Papers on Economic Activity: 1 (Washington: Brookings Institution) pp. 75148.

    • Search Google Scholar
    • Export Citation

    MaretXavier and GerdSchwartz1993“Poland: The Social Safety Net During the Transition,”IMF Working Paper 93/42 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    MaretXavier and GerdSchwartz1994“Poland: Social Protection and the Pension System During the Transition,”International Social Security ReviewVol. 47No. 2 pp. 5169.

    • Search Google Scholar
    • Export Citation

    MilanovicBranko1992a“Income Distribution in Late Socialism: Poland, Hungary, Czechoslovakia, Yugoslavia, and Bulgaria Compared,”Research Project on Income Distribution During the Transition Research Paper Series No. 1 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    MilanovicBranko1992b“Social Costs of Transition to Capitalism: Poland 1990–91,” Research Paper Series, No. 2 (Washington: World Bank,June).

    • Search Google Scholar
    • Export Citation

    MilanovicBranko1992c“Poverty in Poland, 1978–88,”Review of Income and WealthVol. 38 (September) pp. 32940.

    MilanovicBranko1993“Social Costs of Transition to Capitalism: Poland 1990–91,”Policy Research Paper No. WPS 1165 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    OchockiAndrzej1993“Information on Poverty in Poland” (unpublished; Warsaw: Central Statistical Office).

    PanekTomasz and AdamSzulc1991“Income Distribution and Poverty Theory and a Case Study of Poland in the Eighties” (unpublished; Warsaw: Central Statistical Office Research Centre for Statistical and Economic Analysis and Polish Academy of Sciences).

    • Search Google Scholar
    • Export Citation

    SchwartzGerd1994“Public Finances” in Poland: The Path to a Market EconomyIMF Occasional Paper No. 113,ed. byL.Ebrill and others (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    United Nations Children’s Fund (UNICEF)1993Central and Eastern Europe in Transition: Public Policy and Social ConditionsRegional Monitoring Report No. 1 (Florence, Italy: UNICEF, November).

    • Search Google Scholar
    • Export Citation

    World Bank1993Poland: Income Support and the Social Safety Net During the Transition (Washington: World Bank, January).

    Other Resources Citing This Publication