Poverty in Poland

The analysis is based on the 1993 Household Budget Survey (HBS) conducted by the Polish Central Statistical Office. The data are for the first six months of 1993 and are expressed in June 1993 prices. Those who are to be considered to be "poor" are all members of a household whose expenditures per equivalent adult are below the level of the minimum pension. The minimum pension is used in Poland to screen applicants for social assistance (if a household earns more than that level per person it is disqualified for social assistance). In June 1993, the minimum pension was about $71 per month, which, taking the 2:1 ratio between the purchasing power of the Polish zloty and the current exchange rate, equals about $PPP140 (PPP=purchasing power parity or international dollars) per month, or a little over $PPP4 per day per equivalent adult. This is about $PPP3 per person per day.

Using this poverty line, 14.4 percent of the Polish population was classified as poor. The estimated number of poor, however, is very sensitive to small changes in the poverty line. This is because a lot of the population is "bunched" closely around the poverty line. Thus, if we raise the poverty line by some 25 percent, the number of poor almost doubles, rising from 14.4 percent of the population to 26.2 percent. Along the spectrum of conceivable poverty lines ranging from the minimum pension ($PPP4 per day) to the social minimum ($PPP10 per day), each 10 percent increase in the line adds about 6 percent of the population to the ranks of the poor.

Throughout the 1980s, the poverty rate (calculated using the same real poverty line as for 1993) oscillated between 5 and 10 percent of the population. In 1990, the first year of the stabilization, it jumped to 15 percent where it has stayed since. However, as shown in Figure 1, the poverty rate in 1993 seems to be edging downwards.

Poverty can be said to be "deep" if the average income (expenditure) of the poor is far below the poverty line. The poor are then clearly a separate "underclass." This is not the case in Poland. Polish poverty is "shallow." On average, the average poor person's expenditures are only 12-15 percent less than the minimum pension. Shallowness persists with the use of higher poverty lines. It is due to the fairly compressed distribution of expenditures among the poorer segments of the population. This means that few of the poor have extremely low expenditures—in other words, there is no identifiable "underclass."

Who are the poor?

The population can be divided according to socioeconomic group (for example, workers or farmers), education level, gender, type of locality (for example, big city or medium city), and age. Table 1 summarizes who the poor are using several of these criteria. Figures 2 and 3 show that the poverty rate (the percentage of the poor within a given group) uniformly declines with age and education level.

• Over one-third of the poor are poor because somebody in their household is unemployed. About 60 percent are poor because their income is insufficient even if there are no unemployed people in their households. (This 60 percent break down almost evenly between workers, on the one hand, and farmers and mixed on the other.) Only 5 percent of the poor are old.

• About 60 percent of the poor live in villages (40 percent of the Polish population as a whole live in villages). An additional 27 percent live in towns with under 100,000 inhabitants. Therefore, only 13 percent of the poor live in larger cities.

• People who live in villages are not only farmers. In effect, only one-quarter of them are farmers and an additional one-fifth are worker-farmers. Thus, the high share of rural poverty does not directly translate into a high rate of farm poverty.

• People living in large families (with three and more children) account for 30 percent of the poor. However, more than 40 percent of the poor belong to "non-nuclear" families (for example, three-generational families, or a couple with children living with spouse's cousin).

• Since large families are so prominent among the poor, children also account for many of the poor. In effect, one-third of the poor are children under 15 years of age, about 61 percent are people of working age, while only 5 percent are elderly.

Figure 2. As function of age	Figure 3. As function of completed years of education

Incentive and Regulatory Framework.

• Growth. The strong relationship between income and poverty is illustrated in Figure 1. The resumption of growth is crucial to reduce poverty. However, the implication of shallow poverty is that even small increases in income may have a substantial impact on poverty. It is estimated that each percentage point increase in income (without distributional changes) produces a 0.5-0.7 point reduction in poverty. In other words, a 5 percent growth rate should reduce the percentage of the poor from the current 14.4 percent to between 11 and 12 percent in one year. If such a growth rate were to be maintained over the remainder of the decade, the poverty rate could drop to about 5 percent of the population by the year 2000.

• Labor policies. Since 95 percent of the poor are poor because they are unemployed or because their wages are low, labor policies are the most important factor that affects poverty. There are currently several types of "labor interventions" such as public works, wage subsidies for the unemployed, and retraining. Public works are organized by local governments. They concentrate on the maintenance of local infrastructure, construction works, and municipal services. They last up to six months. Public works wages cannot exceed 75 percent of the average wage (this is twice the minimum wage or twice the "flat" unemployment benefit). Wage subsidies are received by employers who create new jobs, that are taken by the unemployed. Employers receive, for a period of up to six months, a subsidy equal to the worker's unemployment benefit. Training programs for the unemployed are fully financed by the government, even when training services are subcontracted to private training agencies. The unemployed receive a "training benefit" that is 15 percent of the average wage in excess of the regular unemployment benefit.

Wage subsidies seem to be more efficient than public works. They are less costly for the government, and workers tend to find "real" jobs afterwards more often. However, employers who receive wage subsidies usually require higher-skilled workers than do the public works. This makes it more difficult to use wage subsidies to employ the long-term unemployed. Neither wage subsidies nor public works seem to be particularly effective at solving the problem of long-term unemployment. This may also be due to the actual unemployability of the hard-core long-term unemployed, resulting, for example, from the complete lack or erosion of their skills.

Experience with training programs varies significantly between voivodships (administrative units in Poland). In some, the impact of training on the chances of finding employment is close to zero. In others, it is estimated that up to one-half of ex-trainees find a job within six months. This variation is explained by the quality of courses offered and by variation in the quality of training agencies and their awareness of the structure of labor demand. The better ones come close to the experience of training programs in the most successful developed countries.

Public Expenditures and Social Safety Net

Social cash transfers, which include pensions, family allowances, unemployment benefits, social assistance and other social insurance payments (for example, maternity leave and sick leave), accounted for 19 percent of GDP in 1993. The single largest component is pensions: 15 percent of GDP. Unemployment benefits account for 1.9 percent of GDP. Cash and in-kind social assistance is small—a little over 0.5 percent of GDP. Overall, social transfers are very costly. Poland spends more—in terms of GDP—than standard market economies at the same or higher levels of income. Portugal and Spain, with per capita GDPs respectively two and three times higher, spend 11 and 13 percent of their GDP on social cash transfers.

About 50 percent of households receive pensions and family allowances. Family allowances cover almost 90 percent of Poland's children (the only exceptions being the children of "rich" farmers because family allowance is income-tested for farmers). Unemployment benefits are received by almost 10 percent of households, and social assistance by 3.6 percent. In total, 68 percent of the population receive at least one social transfer other than pensions. Among the poor, four out of five receive at least one social cash transfer other than pensions (see Figure 4).

Between 50 and 80 percent of the recipients of social transfers (other than pensions) are not poor even before they receive a particular transfer. The worst record is that of the family allowance, 80 percent of whose recipients are non-poor. How much money goes to the non-poor—in other words, how important is the "leakage"? Table 2 shows that leakage ranges from 40 to 74 percent of total transfers. It is significant that even for social assistance, whose key objective is the reduction of poverty, the leakage amounts to 40 percent of spending.

Anti-poverty Strategy

• Labor market policies. Public works should be treated as a measure of last resort and should be strictly targeted to the long-term unemployed and the poor. The main effect of wage subsidies is the advantage they give to the targeted group compared with other job seekers. Such a redistribution of jobs may be justified if it gives employment to disadvantaged groups, who otherwise might never get a job. Wage subsidies should, therefore, be targeted to groups with the least competitive capacity, such as the long-term unemployed and women.
• Social assistance. The current system uses the minimum pension per capita as a screening device for social assistance eligibility. However, it does not guarantee that level of income to anyone. Several elements indicate that a minimum guaranteed income (MIG) should not be introduced at present. First, the costs of such a guarantee would be between 2 and 4 percent of GDP. Second, social assistance offices are already swamped, due to increases in the numbers of applicants in the last few years, and their capacity to assess the income and employment situation of applicants is weak. Third, the incentive effects of a MIG would be negative; there would be practically no difference between the MIG and the minimum wage. Incentives to look for a job would decrease. The poverty trap problem would become sharper. Instead of getting "regular" jobs, people would be pushed toward the informal sector, where they could work but not pay any taxes and hope to be able to claim guaranteed income.

• Workfare. Social workers feel that social assistance recipients, particularly those who have exhausted their unemployment benefits, should participate in community service as condition for receiving aid. This view is based on the fact that incomes and employment in the informal sector are extremely difficult to verify. A significant proportion of people who collect social assistance is thought to work in the informal sector. Workfare may be introduced, initially on a pilot basis, in voivodships where the numbers of long-term unemployed are the highest.

Statistical System

All Polish poverty studies are based on annual Household Budget Surveys, which are of a very good quality. In 1993, some earlier HBS defects were corrected for the first time. The HBS is currently representative of the entire Polish population. The problem is that the coverage of a number of income sources has become weaker than in the past due to the growth of the informal economy. This is the reason that this poverty assessment used expenditures (rather than income) since the extent of underestimation is thought to be smaller with expenditures than with income.

Since there is already an adequate database, it should be relatively easy to introduce a formal poverty monitoring system. The system would have two major components, both updated annually: a poverty profile and information regarding the incidence of social transfers. Based on these two elements, an independent agency, entrusted with the implementation of the monitoring system, would suggest improvements in the delivery of social transfers to the government.