Report No. 26011-YU Serbia and Montenegro Poverty Assessment (In Two Volumes) Volume I: Executive Summary November 13, 2003 Poverty Reduction and Economic Management Unit Europe and Central Asia Region Document of the World Bank VOLUME ONE: TABLE OF CONTENTS ACKNOWLEDGEMENTS ....................................................................................................................... i ABSTRACT ................................................................................................................................................ ... 111 EXECUTIVE SUMMARY ........................................................................................................................ 1 EconomicReformsAnd Poverty..................................................................................................... The PRSPAgenda And Background............................................................................................... 2 What I s Poverty InSAM?................................................................................................................ 5 3 How Many Are Poor?...................................................................................................................... 7 Who Are The Poor?....................................................................................................................... 10 Gender And Poverty...................................................................................................................... The PoorOnThe LaborMarket.................................................................................................... 12 14 ReducingVulnerability: Safety Net And The Poor....................................................................... The Challenge Of RehabilitatingHumanCapital.......................................................................... 17 RegionalContext........................................................................................................................... 18 DataFor Policies............................................................................................................................ 26 Strategy for PovertyReduction..................................................................................................... 22 22 Recommendations.......................................................................................................................... 28 Conclusions.................................................................................................................................... 31 MATRIX OFTHE REPORT .................................................................................................................. 32 REFERENCES ......................................................................................................................................... 39 GLOSSARYOFTERMS ......................................................................................................................... 41 Listof Figures Figure 1:SAM: Changeinthe PovertyHeadcount 1990-1999....................................................... Figure3: PovertyRateRelativeto the Average by Gender of HouseholdHeadinSerbia........... 13 Figure2: Material andNon-incomePoverty................................................................................... 4 9 Figure4: LaborForceParticipationRates inS A M . 1995-2002................................................... 15 Figure5: Coverage. TargetingandAdequacy of SocialProtectionPrograms in Serbia..............20 Figure6: Distribution of Region'sPoorby Countries.Around 2000-02....................................... 22 Listof Tables 7 Table 2: PovertyamongAdults Accordingto EducationinSerbiain2002................................. Table 1: SerbiaandMontenegro: Material Poverty in2002........................................................ 10 Table 3: PovertyAccordingto the Employmentof HouseholdMembers inSerbiain 11 Table 4: PovertyProjectionsfor Serbia. 2002-06......................................................................... 2002 ............................................................................................................................................... 25 List of Boxes Box A: Programmatic Approachto PovertyWork inS A M........................................................... 2 ACKNOWLEDGMENTS This Poverty Assessment report i s the first output of the poverty work program conducted in collaboration between Serbian and Montenegrin counterparts and the World Bank. On the Serbian side the team was led by H.E. Gordana Matkovic, the Minister of Social Affairs of Serbia, who played a prominent role in the successful implementation of the household survey project, the analysis of data and the dissemination of results. Mrs. Snezana Mijuskovic, Deputy Minister of Labor and Social Affairs of Montenegro and the Head of PRSP Management Unit was the key Government counterpart in Montenegro. On the World Bank side the team was coordinated by Ruslan Yemtsov (Senior Economist, ECSPE). Data for this report were generated with the support of the Poverty Survey Project (No. P074904), co-financedby the Dutch Trust Fund. In Serbia, The Ministry of Social Affairs commissioned the survey in 2002 to feed into the social policy reform program and the PRSP. The survey was conducted by SMMRI (Belgrade) in May-June 2002. The ISSP-CEED (Podgorica) headed by Petar Ivanovic, achieved a similar result inMontenegro inAugust-November 2002. Our Serbian and Montenegrin counterparts initiated the analysis of data with pioneering papers by Gorana Krstic (poverty profile for Serbia), Bosko Mijatovic (social policy assessment and reform), Biljana Bogivevic (social protection), Branko Milanovic (inequality and social assistance analysis), Vladimir Vukojevic (social impact analysis) and Dragana Radevic (poverty in Montenegro). The team i s indebted to many government officials from Serbia and Montenegro ( S A M ) . The team also heldnumerous consultations with the PRSP working groups and benefited from discussions with various international organizations, including UNICEF, UNDP, EAR and DFID. Key analytical inputsto the report were prepared as backgroundpapers by CemMete (health and poverty, education for poverty reduction), Cornelia Tesliuk (social protection), Kathleen Beegle and Dragana Radevic (Poverty in Montenegro) and Lazar Sestovic (macro policies). Jim Stevens (education), Loraine Hawkins (health), Mamta Murthi(social protection), Taies Nezam(IDPs and refugees), Henry Gordon (rural issues), Branko Jovanovic and Mihail Arandarenko (labor market) provided inputs. Emily Evershed was an editor of the report, Helena Makarenko, Judy Wiltshire and Jim Lynch undertook the report processing. Rory O'Sullivan and Marina Petrovic from the World Bank Belgrade office provided a link to policy dialogue. Zana Ivanovic seamlessly organized numerous missions and timely translations required for fully collaborative work. The team i s wholeheartedly grateful to Nancy Cooke (Lead Country Officer for S A M ) for her support and encouragement. Asad Alam (Poverty and Gender Sector Manager, ECSPE) guided the entire process of preparing the report and provided crucial advice in conceptualizing the report and in finalizing its production. The report was prepared under the guidance of Orsalia Kalantzopoulos (Country Director, ECCU4). The peer reviewers were Jeni Klugman (Lead Economist, PRMPR) and Salman Zaidi (Senior Economist, SASPR), who contributed substantively to the report's analysis and presentation. i .. 11 ABSTRACT The Poverty Assessment i s the first public service provision. Even though output of a multi-year program adopted some of the social assistance programs, by the World Bank to assist the such as MOP in Serbia (Materijalno Governments of Serbia and Montenegro Obezbedenje Porodice), are among the in the development and implementation best targeted programs in the region, the of their Poverty Reduction Strategies. social protection system as a whole The program relies on collaboration in suffers from large exclusion errors. joint data production and analysis. Key findings of the report represent a Given the high level of vulnerability of consensus view on poverty achieved in the population and the shallowness of the process of collaboration with the poverty, a broad-based growth strategy Serbian andMontenegrin Authorities. that ensures that the benefits accrue at least proportionately to the poor i s Based on data collected in 2002, the central for acceleratedpoverty reduction. report finds that absolute material Improvements in the business climate poverty affects every tenth person in will stimulate private sector growth and both Serbia and Montenegro. For feed into employment generation - a SAM's historical standards this is a very necessarycondition for pro-poor growth. high incidence. Inequality remained Growth will increase fiscal revenues to moderate by regional standards, and as a remedy the problems of chronic under result poverty i s shallow with the funding, while structural and public poverty depth equivalent to around 1 administration reforms will strengthen percent of GDP. At the same time the governance and the quality of vulnerability--or exposure to negative services provided to the poor. The shocks and inability to cope with them-- multidimensional nature of poverty threatens many currently non-poor requires concerted and well coordinated individuals. At least as many suffer from action in different sectors. The report deprivation in other dimensions of well argues for special programs to address being, such as health, education, the needs of socially excluded. PRSP housing, social inclusion or property requires institutionalization of poverty rights. Material poverty, therefore, is not monitoring as a function of the statistical the only challenge for the Governments. system. Four factors are most strongly related to The report is organized in two volumes. poverty: low education attainment; Volume One (Executive summary) joblessness; the location in rural areas summarizes the Report content. Volume and depressed regions, and the presence Two (Main report) provides detailed of socially disadvantagedmembers (such results of poverty analysis. Due to data as IDPs or Roma). The poor are found limitations sectoral part of the main to face serious problems of access to report covers Serbia in greater details. public services (health, education, An analysis of available data for sanitation) and suffer disproportionately Montenegro i s presented in a from the deterioration in the quality of background paper. ... 111 EXECUTIVE SUMMARY 1. There are about 2.2 million people living in absolute poverty in all countries and terrotories that once formed the single state of the Socialist Federal Republic of Yugoslavia (SFRY). About 41 percent of these people live in Serbia and Montenegro (SAM).' This high share of the poor i s a function not only of the larger share size of the population in SAMbut also of country's incidence of poverty, which i s only next to that in Kosovo and Bosnia and Herzegovina. This report attempts to answer the question of how a country that enjoyed high standard of living at the dawn of the transition, has become impoverished relative to its neighbors, and to assess the implications of this situation. 2. This Poverty Assessment is the first output of a multi-year program of analytical, monitoring and capacity building poverty work adopted by the World Bank to assist the Governments of SAM in the development and implementation of their Poverty Reduction Strategies. The report was preceded by a two-year period of intensive cooperation and joint analysis of data collected by the Survey of Living Standards (SLS) in Serbia (commissioned by the Ministry of Social Affairs with the support of the Dutch Trust Fund to the SMMRI, a local research firm), and the Household Survey (HHS) in Montenegro conducted by ISSP-CEED (a local NGO supportedby the EC Food Security Program and USAID). The first roundof surveys was completed in 2002, generating baseline data for national Poverty Reduction Strategies and this report. Subsequentroundof data collectionis on-going. 3. This baseline Poverty Assessment proposes key policy conclusions and priorities for public action based upon the jointly developed detailed poverty diagnostics, and identifies key knowledge gaps that remain to be filled. As the situation in SAMi s changing rapidly, this report will be followedby additional data collection and analytical work (See Box A). 'IfKosovoi s poor inside the borders of S A M are counted, this proportion rises to 66 percent. 1 BOX A: ProgrammaticApproachto Poverty Work inSAM. The Poverty Assessment i s the first product in a series of tangible outputs: analytical reports, poverty maps, information support for the national strategies for poverty reduction, series of background papers on key policy topics (labor, inequality, rural poverty, etc.). These outputs will be strategically timed to respond to the needs o f the Governments. Inparticular additional efforts will be made toward further analysis to fully understand the causes of poverty, with' a focus on labor market, its regional characteristics and ethnic dimensions, such as poverty situation of ethnic minorities (with an emphasis on Roma). These would also contribute to the future development and effectiveness of the Bank's Transitional Support Strategy (TSS) and design of various operations. This approach focuses onjoint analysis, capacity building and continuous disseminationof results. At the next stage of the Poverty Assessment work the analysis will directly engage Statistical offices o f both Republics, and a series of skills-focused training course will be offered to data producers and users incooperation with WBI. The proposed set of key poverty analytical outputs for S A M for the medium-term is given below. This program will be revised and updated over time to reflect changing country situation and demands. Outputsof The PovertyAssessmentProgram. Key activities and outputs FY02 FY03 FY04 FY05 Poverty Survey J J PovertyAssessment:BaselineReport J J Background Paper: Poverty Profile for Montenegro J Background Paper: Evaluation of Social Assistance System inSerbia J Background Paper: Monitoring Poverty inSerbia (Building a Statistical System) J Background Paper: Chronic Poverty and Social Exclusion inS A M J Background Paper: Rural Poverty inSerbia J Background Paper: Private Sector Development, Labor Market and Poverty inS A M J Poverty Map for S A M J Background Paper: Enterprise Restructuring and Unemployment in S A M J Background Paper: Options for Pension Reform and Poverty inMontenegro J Background Paper: Vulnerability to Poverty and Health Care Reform Issues inSerbia J Background Paper: Education Reform and Poverty Reduction in S A M J PovertyUpdateReport J Note: FY -fiscal year Le. FY 03 is July 2002-June2003. This program of work has been initiated by the joint World Bank-Government team in consultations with UNDP, UNICEF, DFID, SIDA, CRS, IFRC and Yugoslav Red Cross, WHO, UN WFP and UN FAO, European Commission Food Security Program and European Reconstruction Agency as well as with local researchers and NGOs in Serbia and Montenegro as partners in the development and implementation of the poverty reduction strategy. The work is closely coordinated with several on-going and planned operations and analytical studies undertaken by the World Bank: (i)TSS discussions and finalization, (ii) Social protection reform projects; (iii) Serbian Structural Adjustment Credit 11; (iv) Health and education sectors reform work; (v) Public expenditure review (PER); (vi) Labor Leaning and Innovation Loan; (viii) Country Economic Memorandum. ThePRSPAgenda and Background 4. The Governments of Serbia and Montenegro are keenly aware of the issue o f poverty and associated social tensions. The social issues were in the forefront o f change in Serbia (far- reaching pension reforms, the re-targeting of child benefits, programs to assist employment restructuring). Pension system reform and social policies were also among the key policy issues inMontenegro. But most of these reforms to date are being driven by the fiscal unsustainability 2 of old system and the need to keep social entitlements within manageable fiscal limits. Thus, there i s an urgency inunderstanding the social andpoverty impacts of ongoing reform. 5. The poverty focus was strengthenedwith the initiation of the PRSP inboth Republics in early 2002 in an open, participatory fashion. The country completed an Interim Poverty Reduction Strategy Paper (I-PRSP) in July 2002. A series of broad-based consultations over 2002-03, and significant analytical and organizational efforts, led to the preparation of the draft of the full PRSP. The first drafts of the full PRSP were made public for consultations in Spring 2003, and final versions were adoptedby the Governments in October-November 2003, 6. The paucity of data on poverty was a key stumbling block identified by the Transition Support Strategy of the World Bank dating back to 2001 (World Bank [2001a]), by the I-PRSP and by the Joint Staff Assessment of the I-PRSP (IMF/IDA [2002]). The poverty assessment itself and the preparatory work inproducing the data required for the assessment (Poverty Survey TA Project) fill this critical gap. This assessment usedthe most recent data available for Serbia and Montenegro and stimulated a participatory process to analyze them, so as to arrive at a shareddiagnosis of poverty. 7. This report was producedby ajoint team comprising localexperts andthe Bank staff (see Acknowledgments section); it represents a consensus view on poverty in SAM. The highcaliber of experts involved in the analysis on both sides ensured both the high quality of the diagnostics work and the status of both sides as equal partners. The key product of this joint team was the creation of a single data platform with the PRSP working group. This commonality of the data base means that a number of key conclusions presented in this report are fully reflected in the PRSP documents. Economic Reforms and Poverty 8. The beginningof 1990swere the hardest years for the FRY inits short history owing to a number of internal and external political and economic shocks. The problems were accentuated by the pursuit of poor economic and social policies. These led to a deep and sharp output decline (to less than half of its pre-transition level), accompanied by hyperinflation, a rise in 3 unemployment, a huge increase in internal and external public debt, and an increase in the informal sector. The impact on the living standards was obvious - poverty increased significantly, as Figure 1shows. Figure 1: SAM: Change inthe Poverty Headcount 1990-1999 (1990=100%) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Nore: Usingthe FSO foodbasketandHBS data onhouseholdincomes; without Kosovo; dataincludebothSerbiaandMontenegro Sources: Posarac (1997) and B. BogiCeviC et al. (2001). 9. InOctober 2000, a change of government in Serbia2markeda turning point in the post- socialist history of the country. Reversing the negative trends of the past, an ambitious program of economic reforms was initiated with donor su~port.~Future integration of the economy with the European Union provides a new anchor for policy. Monetary and fiscal policies have made a markedbreak with the past, achieving some measure of macroeconomic control (with inflation down to 14.2 percent in 2002 from 113.5 percent in 2000, and with the budget deficit within manageable limits). Major tax reforms were implemented, and wide-reaching reforms were startedin some sectors (energy, social protection, labor market, banking, etc.). 10. The GDP grew by some 5 percent per year in 2000 and 2001, slowing somewhat to 4 percent in 2002. Growing wages and householdincomes have stimulated increasedconsumption (which outpaced GDP growth), but little output response, as demands have largely been met by imports. Over 2001-02 there was little, if any, positive employment response, suggesting significant slack in capacity utilization. The labor force survey unemployment rate even A real change of power occurred when the previous regime first lost elections and then accepted election results owing to massive street protests. Reforms in Montenegro started earlier and were accompanied by strife for independence. Based on this programthe World Bank formulated its own country support program - TSS (World Bank [2001a]). 4 increased slightly, from 14 percent in2000 to 15 percent by 200L4 This has colored perceptions of the populationas monitored through opinion polls. 11. New policies were implemented with little information available about the characteristics of the poor. From the start, reformers had highlighted their objective of promoting human welfare and life with dignity. Opponents of reforms had painted terrifying pictures of the population inmisery and social unrest inresponse to any attempt at reform. 12. This report provides an objective assessment of poverty. It shows that most gloomy pictures are far from the reality. It also confirms that the focus of the reform strategy -- on sustainability, institutional reforms and an improved business environment -- i s fully warranted bythe accumulated domestic imbalances that harmfuture prospects for poverty reduction. WhatI s Poverty In SAM? 13. The report i s based on a multidimensional characterization of poverty. Material absolute poverty is defined as lack of consumption of essential food and non-food goods and services. The poverty lines for Montenegro and Serbia were established in ajoint analytical work with our counterparts (Krstic in Bogicevic et a1 [2003], Beegle and Radevic [2003]). The decision while establishing these lines was to follow the actual consumption patterns of the population andto basethem on objective evidence, not onjudgment. 14. To anchor absolute poverty lines in the economic reality of each country, the approach was to take local cost of the minimum food basket that meets key nutritional requirements and follows the consumption patterns of the populations. This poverty line i s used to identify the extremely poor. 15. To derive the full, baseline poverty line, the survey data were usedto find out, based on actual consumption behavior, at which level of total consumption such minimum food Registeredunemploymentgrew evenfaster, from 26 to 29 percentbetween 2000 and 2002, but as shown inthis report registration statisticsare misleading. All former SFRY republics are known to have large discrepancies betweensurvey and registeredunemployment rates which are thought to reflect the incentives createdfor registration by the designof the social benefit system. 5 requirements of all household members will be met.6 This poverty line has a very clear meaning: all households with total consumption below the poverty line suffer from material deprivation, as either they lack adequate nutrition, or, to meet basic food requirements they must cut back on other essential needs. Such an assessment of the poverty threshold for mid-2002 yields a level of 2.40 a day in Serbia (at market exchange rate) and 3.50 a day inMontenegro as poverty lines. 16. To assess the vulnerability to poverty one needs panel data that follow same households over time. In the absence of panel data in SAM, the approach was to use two arbitrary cut-offs. The first, adopted by the Serbian PRSP team, was to set it at the level of the upper bound of the poorest quintile of the population. It was used to assess the characteristics of near-poor households. B y definition, the incidence of vulnerability with this approach i s always 20 percent. The second approach adopted in this report for Serbia and Montenegro was to set a vulnerability line that i s 50 percent above the full poverty line.7 17. To assess whether a household i s poor or not, its current level of consumption per equivalent unit i s compared to the poverty line. Consumption includes purchases, money value of in-kind transfers and own production of non-durable goods and services, imputed rent for owner-occupied housing, and flow of services from other durable assets owned by a household. 18. Non-income poverty i s measuredwith the household survey data in several dimensions of well-being: education poverty, health poverty, housing poverty and employment poverty. In each dimension a minimum standard i s established reflecting the level of development achieved in SAM. For example, education poor are defined as adults with only uncompleted primary schooling or less (in other countries education poverty may be definedas illiteracy), and housing poverty i s defined as substandard housing (households not connected to tapped water and using latrines, households living in a building unsuitable as dwelling or living in overcrowded dwellings with more than 3 persons per room). To the extent it makes sense, these definitions ~ ~ ~ ~ Usingthe World Bank methodology as outlined inRavallion (1992), food poverty line is comparedto total per capita consumptionto estimateextreme poverty. As discussed inRavallion(1992), this approachestablishes a consistentabsoluteupper-boundpoverty line; as non- food costs represent a significant share of consumption inS A M , even for the poor, the variant to set the lower-bound line wasjudged to be not applicable. 'Suchpracticei s often used inthe poverty analysis (see PRSP Source Book). 6 are made comparable to the set of indicators recommended by the Eurostat task force on social exclusion (see Atkinson et al. [2002]) and MillenniumDevelopment Goals. How Many Are Poor? 19. Around 10 percent of the population in Serbia and Montenegro were in absolute material poverty in mid-2002 (see Table 1). The assessment also finds some extreme poverty, which has been eradicated in all but two countries inthis region (Kosovo and Albania). * Table 1: Serbia and Montenegro : Estimates of Material Poverty in2002. (Percent of Podation) Extreme Absolute Vulnerable to (baseline) poverty or poor Serbia 2.3 10.6 32.6 Montenegro 0.0 9.4 36.4 Source: SLS 2002 for Serbia and ISSP-CEEDHHS2002 for Montenegro Nom: Extreme poverty is definedas current consumptionper capita without imputedhousingrent below the cost of the food basket, absolute poverty i s defined as full consumption below the cost of the poverty line, and the vulnerability line i s set at the level 50 percent above the poverty line. See Krstic (2003), Beegle and Radevic (2003) for detailed definitions. The data presentedinthe table do not include information on certain vulnerable groups (such as Roma) or only partially cover them and therefore are biased downward. 20. Figures presented in the Table 1 must be taken as estimates o f poverty. First, certain groups are not covered by the household survey by design (institutionalized population). Such groups, for example, IDPs in collective centers, may be particularly poor. Second, some groups are extremely hard to reach in a general population survey. Different groups of Roma minority is a case in point, as well as some IDPs in private accommodations. New data for Montenegro demonstrated that the poverty rate for the entire country may be revised upwardby as much as a fifth (from 9.4 percent to 12.2 percent), once these IDPs, refugees and Roma are fully incorporatedin the survey sample (ISSP-UNDP [2003], see Annex Ito Vol. I1for details). These most recent data also show the presence of extreme poverty in Montenegro. The recent SLS data collection round (2003) in Serbia included a special survey of Roma, and its results need to be carefully assessed to arrive at a more accurate poverty estimate for Serbia. Fourth, numbers from any survey have a certain degree o f precision. The statistical confidence interval around * Theextremepoverty estimatereported inTable 1is basedon the World Bank international methodology different from baselinenationally-specific approaches followed inSAM; the methodology is basedon the insights frompoor countries, where significant number of extremely poor allow an in-depthanalysis of measurementassumptions. 7 the measured poverty rate for the SLS 2002 is +1.2 percentage points and for H H S in Montenegro it i s +1.9 percentage points. Table 1 reports the exact figure; it would be more accurate to say, for example, that poverty in Serbia i s estimated to be between 9.4 and 11.8 percent. This may affect the extreme poverty figures particularly strongly and call for careful use and interpretation of results. But even 9 percent for absolute material deprivation incidence produced with the data for the year 2002 tells that poverty is a very serious problem. 21. To assess the extent of poverty, one also measures its depth andseverity. The poverty gap using the baseline definition of poverty in Serbia i s equal to 2.2 percent and in Montenegro is equal to 1.3 percent; it means that the consumption of apoor person in Serbia on average falls 20 percent short of the poverty line (14 percent in Montenegro). Poverty therefore i s shallow (especially in Montenegro), and the total poverty gap i s around 1percent of GDP. This i s a reflection of moderate inequality (Gini index for consumption i s around 0.3 - see Milanovic in Bogicevic et a1 [2003]). 22. As poverty is shallow it is very sensitive to the changes of the poverty line or incomes. More than 20 percent of the population in Serbia and over 25 percent inMontenegro have a consumption that i s "just" above the poverty line, "just" meaning a maximum of 50 percent above the line. Increasing the poverty line by 10 percent raises poverty at both the household and individual levels by approximately one-third. Decreasing the poverty line by 10 percent lowers poverty at both the household and individual levels also by about 30 percent. This suggests that even small economic shocks can have potentially large effects on poverty. A positive shock (such as those associated with growth and good policies, for example) i s associated with more-than-proportional declines in poverty; a negative shock (such as a recession) i s associatedwith more-than-proportional increases inpoverty. 23. Poverty in other dimensions (housing, health, education, employment, as aspect of social inclusion) is not negligible. Although the picture i s mixed, the multidimensional indicators of poverty can be combined to provide further insights into the characteristics of the poor and the causes of poverty: 0 Every sixth adult inbothSerbia and Montenegro can be considered "education poor"; 0 Healthpoverty affects between4 and 6 percent of the population in SAM; 8 0 Housing conditions for 10-15 percent o f the population in both Republics are below the poverty standard; in addition, as many as 5 percent do not enjoy clearly established rights on their housing; 0 Over 3 percent of working age adults in Serbia are willing to work, available to work but were outside of work for more than two years continuously, and therefore fail in social inclusion. 24. On a more positive side, simultaneous deprivation in several dimensions i s rare. This confirms that extreme poverty in the country i s not widespread. But as poverty consists of non- overlapping, or only partially overlapping, groups (see Figure 2), a significant share o f the population i s poor at least in one dimension. As many as 30 percent of all adults are poor at least inone of the poverty dimensions: materialconsumption, education, health, employment, housing or property rights. This higher proportion o f those affected by various aspects o f poverty probably feeds the popular perception of widespread poverty inthe two Republics. Figure 2: Material and Non-incomePoverty n Note: size and overlap of circles are suggestive of the data on the number of people affected. 25. Material poverty, therefore, i s not the only challenge, and i s probably not the most daunting challenge, for the Governments in SAM. Many more people are found to be near-poor and in a fragile state just above the poverty line, and they are vulnerable to any economic fluctuation, downturn, or personal shock. Many people suffer from non-income forms of deprivation. 26. This diagnosis leads to the first conclusion: reducing multi-dimensional poverty in SAM requires coordinated action in many sectors. While economic growth i s central to reduce both material poverty and vulnerability, other aspects o f poverty have to be addressed simultaneously. 9 WhoAre the Poor? 27. Four variables are most strongly related to poverty: these are education attainment; employment status; the location of the household; and the belonging to one of socially disadvantaged groups. Recognizingthe differential poverty risk for different groups of the population can provide useful insights into which factors are associated with poverty, and thus helpdesignpotentialpolicyresponses. 28. First, the strongest and closest correlate of poverty is education. Poorly educated individuals make up the majority o f the poor and have the highest poverty risk. They are also more likely to face problems of discrimination in the labor market. Table 2 shows the poverty rates by level o f education inSerbia (very similar outcomes are observed inMontenegro). Table 2: Poverty amongAdults Accordingto EducationinSerbia in2002 (populationover 15 years of age, percent) Extremely poor, % of Poor, % of Share inthe amongthe Share Poverty Poverty population population population poor depth severity Uncompleted elementary school 4.1 21.5 17.5 35.4 5.1 1.8 Elementary school 3.1 14.3 23.3 31.2 3.1 1.o Secondary school 1.6 6.9 47.5 30.8 1.2 0.3 Jr. college 1.3 2.9 5.1 1.4 0.4 0.1 University 1.2 2.0 6.6 1.2 0.4 0.1 Total for adults 2.3 10.6 100 100 2.2 0.7 Source: Krstic inBogicevicet al. (2003) and staff estimatesbased on SLS (2002). 29. Multivariate analysis o f poverty risks suggests that lower education attainment i s the key background factor explaining the higher incidence o f poverty of some vulnerable groups, such as rural households orjobless households. For example, population inrural areas has a significantly lower educational attainment: 60 percent o f working age adults inrural areas have not progressed beyond vocational school, as opposed to only 30 percent in urban areas. Once the effects o f the lower education in rural areas are controlled for, the estimated poverty risks are equalized between urban and rural households. 10 30. Second, inactivity andjoblessness9are strongly correlated with poverty though the poor can be found in working families too. The absence of working members in working age households more than doubles their poverty rate, as Table 3 suggests (similar results are reported for Montenegro). But, equally important, not all jobless households are poor, and poverty i s not only the lack of employment. The majority of the poor in Serbia -74.6 percent -- are in working families. Table 3: Poverty According to the Employment of Household Members inSerbia in2002 (percent of population) Extremely Poor, % of poor, % of population in a Share inthe among the Share Poverty Poverty population group population poor depth severity Non-active age household 2.4 15.4 8.1 11.8 3.26 1.11 Jobless household 7.5 23.I 6.2 13.4 6.29 2.37 Working household 1.9 9.2 85.7 74.6 1.85 0.60 Total 2.3 10.6 100 100 2.2 0.80 Note: "Working" categoriesare defined on the basisof the ILO standarddefinitions. Source: Staff estimatesbasedon SLS 2002. Definitions are fromEurostat (2003). 31. Working families are in poverty because each working adult in a poor household i s earning significantly less than an adult from a better off group. Why? Partly because the poor live in the "wrong" places - regions affected by closure of enterprises, lay-offs, and a poor business climate, but mostly because their lack of skills makes them less productive. Therefore this disadvantagestarts with the unequal access to education opportunities. 32. Third, inSerbia andMontenegro, regional disparities are found to be large. These disparities are most clearly reflected in the large differences in poverty rates across regions, It i s striking that a country as compact as Montenegro had 1:2 differences in poverty rates between the poorest and the richest regions. It i s also important that the MontenegrinNorth i s not only the poorest region of Montenegro, but i s also the home of 54 percent of the country's poor. As the well being of individuals depends not only on their own labor market positionbut more broadly on the degree of contactof other householdmemberswith the world of work, the employment profile has to use accurate identification of the labor market status with the household survey as suggested by the definitions of "jobless households" proposedby the EuropeanCommission. The report on indicators for social inclusion preparedby the Social Protection Committee and endorsedinLaeken can be found on the web-site of Directorate General Employment and Social Affairs of the EuropeanCommission (www.europa.eu.int). 11 Regional differences inpoverty rates in Serbia are even greater, exceeding 1:3 between better off Belgrade and the poorest rural areas of South Eastern Serbia. 33. Regional differences in poverty rates cannot be fully accounted for by labor market, education and demographic factors, as the multivariate analysis using the model of household consumption shows. This suggests that there are deeply entrenched disadvantages in certain regions that are likely to have a lasting impact on their poverty, such as the state of social capital inthese localities, accessto infrastructure andto publicservices. 34. Demographic characteristics of a household, including the age profile, number of children and family size, are important determinants of poverty risk under the baseline methodology, but their influence i s very sensitive to even minor changes in measurement assumptions. 35. Refugees, IDPs, disabled and Roma have significantly higher poverty rates. War has also been an important factor affecting the higher poverty among refugees and IDPs. The presence of IDPs and refugees in a household significantly increases its risk to be poor. Same situation i s observed for disabled. The ISSP survey in Montenegro found that the poverty rate among Roma exceeds 50 percent (using the same baseline definition), this i s five times the average incidence (ISSP-UNDP [2003]). The on-going analysis of Roma survey by M O S A in Serbia will provide robust assessment of poverty for this minority. The existing data suggest significant problems of Roma families, especially IDPs, inaccessing education. 36. Only a minority among poor households in S A M is characterized by a combination of four risk factors listed above. This analysis offers insights on how the policies can help the poor. Broad-based multi-sectoral programs combined with better targeting o f focused programs helping the most unprivileged is likely to have an impact on all layers of the poor population. Gender and Poverty 37. The repod shows that in both Republics, female headed households have higher poverty risk. This contradicts the general perception of gender equity in SAM. Using multivariate techniques, the report also finds that for Serbia the observed slight disadvantage o f female- 12 headed households can be fully accounted by a set of economic disadvantages that women face inthe society. The summary of the results is given inFigure3. 38. Figure 3 shows that a female-headed household faces a slightly higher observed poverty rate than does a male-headed household (by 14 percent). But once controls are introduced for demographics and location (region of residence, age profile and household size), one finds that this disadvantage disappears. It means that similar households headed by males also have much higher poverty risks than average. Thus, women are over-represented in households with unfavorable demographic structure (e.g., high dependency rate) located in depressed regions (rural areas). Thus, the observed higher rate o f poverty among female headed households i s in some part due to their specific demographic composition and unfavorable location. After controlling for human capital characteristics andemployment status, it becomes clear that among these factors o f disadvantage, education plays the most important role. 39. The disadvantage that women in S A M face i s apparent in the labor market, as studies of gender gap in pay suggest. The disadvantage of female-headed households also spans the full spectrum of multidimensional poverty measures developed for this report. Figure 3: Poverty Rate Relative to the Average Poverty Rate by of Household Head Gender inSerbia (Higher or Lower Poverty Comparedto Average, Observedversus Simulated) Controlling Observed for Controlling Controlling poverty demographics for Controlling for all of the ratehational andlocation employment for education above t208 -15% 4 1 0 % -g+5% - L. L.4% .- B -t -5cib =-lo% -15% -20% Male headed 0Femaleheaded Source: Staff estimates basedon SLS. 13 ThePoor On The Labor Market 40. Labor is often the only asset a poor household has. Majority o f the poor in S A M are in families where someone works. The situation in the labor market i s therefore a key factor influencing the poor. 41. While assessing the situation of the labor market policymakers in S A M have relied on a limited range o f indicators, overly focusing on the unemployment data provided by the Employment Bureaus. The report argues that this source gives inadequate reflection for the trends on the labor market. The regular Labor Force Survey (LFS) i s by far more accurate and can be used to monitor changes in employment and unemployment according to the internationally accepted E O standards; but LFS has not been exploited for policy purposes. 42. Using LFS data to arrive at more internationally comparable data presents a very different picture to the information from the register. The survey suggests unemployment levels ina range 10-15 percent for SAM" comparedwith a registered unemployment level of over 30 percent. Unemployment i s higher inMontenegro (above 20 percent of the labor force).The main household surveys used in this report (SLS and HHS) confirm the diagnosis based upon the LFS. Most of the registered unemployed are either economically inactive or work in the gray economy. At the same time, almost a third of the "true" unemployed are not registered by the labor offices. 43. LFS data comparable across time show falling labor force participation rates. Figure 4 provides a picture of the changes in the participation rates since 1995. Figure 4 confirms that the i key flow inthe labor market over the entire period was a flow of workers -both men and women - out of economic activity. Women in Montenegro, characterized by low labor market activity rates, were affected particularly strongly in the last two years. The figure also highlights significant gender and regional differences in activity rates. Despite GDP growth by some 5-4 percent per year in 2000-2002, there was no employment response, according to the Labor Force Survey. Overall, there are no signs yet of sizeable gains to employment as a result o f reforms. loWith a range accounting for survey precision and modifications of basic definitions to fit SAM realities. 14 44. This apparent paradox of no employment response to economic growth should not be taken as a surprise. Over the whole period of 1990-2000 significant imbalances have accumulated, due to the reluctance to openly lay-off redundant workers. As a result, despite the fall in output by over 50 percent in 1990-2000, the aggregate employment ( L O definition) fell by only 25 percent leading to a fall in productivity and a rising employment overhang. Consequently, the recent growth has utilized the excess employed rather than create new job opportunities. Figure4: Labor Forces ParticipationRatesinSAM, 1995-2002 (by gender and Republic, for population over 15 years, percent) 75% - 70% Male: Serbia 65% 60% 0 Female:Serbia 55% 50% +Male: Montenegro 45% - 40% A -Female: Montenegro 35% 1995 1996 1997 1998 1999 2000 2001 2002 Source: Staff estimates from LFS (1995-2002). 45. The presence of significant residual excess employment (or hidden unemployment) means that the labor market indicators may show worsening before getting better. The analysis of labor adjustment practiced by firms in the report reveals that a significant share of workers (around 10 percent) were in various forms of reduced work. Their response to beingnot paidon time or being put on forced leave or shorter work schedules varied depending on the type of hours or pay reduction. About a third of them have already found alternative gainful employment, but those who did not faced an imminent threat of becoming unemployed, as they often are characterizedby unmarketable skills. 46. These data suggest that it i s not unemployment per se which i s a key problem of labor market in SAM, but lack of employment and poor use of labor force. A large share of the working age population that i s inactive or occupying themselves with low productivity, 15 occasional work i s a cause of poverty. For the economy-wide perspective this represents output foregone and income lost. Moreover, the important goal of sharing the benefits of growth more equitably among groups and regions proves increasingly elusive when many citizens are trapped inlong-term economic inactivity. 47. The relatively low level of "real" unemployment should not therefore be taken as a source of complacency. The analysis of social assistance reveals that there i s no effective public safety net to protect workers against the risk of unemployment. At the same time, relatively low levels of unemployment do not mean that the labor market in S A M performs well. Behind relatively stable L O unemployment rates, there were important structural changes with significant increases of unemployment rates for new entrants and pre-retirement workers. 48. One o f the viable strategies for laid-off workers in other transition economies has been self-employment and own micro-enterprise start-ups (IFC 2000). These forms of employment outside agriculture are extremely underdeveloped in Serbia and accounted in 2002 for less than 2 percent o f all non-agricultural employment (with a higher share in Montenegro, but still falling below that in more dynamic economies). Even special programs designed to stimulate micro enterprise formation (severance pay schemes) so far have not achieved any impact. 49. Based on the experience of other transition economies," the main impulse for employment generation will come from the sector o f new, small and medium enterprises, whose development critically depends on business environment and restructuring o f formerly state owned enterprises. The hostile business environment hurts the poor directly by limiting the coping strategies available to them (such as self-employment), and indirectly through a weak demand for their labor. 50. The prospects of the poor competing for salaried jobs are hampered by two barriers: the labor market discrimination against the most vulnerable (disabled, IDPs and refugees, Roma, women), and the inadequate skills of the poor. l1See, for instance, World Bank (2002b). 16 51. To make growth pro-poor and to make the labor market work for the poor, it i s important therefore to "pull the strings" that seem to be outside of the immediate realm of labor market intervention: (i) improve the business environment and the rule of law, and (ii) rehabilitate to to the human capital of the poor. An analysis of actions needed for improving the investment climate goes beyond the scope of this report and will be addressed in forthcoming analytical work (Country Economic Memorandum for SAM and Investment Climate Assessment study). Human capital issues are discussedbelow. The Challengeof RehabilitatingHuman Capital 52. Serbia spends less than most other transition countries on education, -- in 2001, less than 3 percent of GDP; this item was twice as high in Montenegro (The World Bank [2002c]). And what was spent was not well spent in the sense that scarce public resources were used to subsidize those who can affordpaying for education themselves, and little flew to the poor. 53. The report finds that children from poor families in SAM are disadvantaged because of a selective education systemthey have a greater tendency to drop out after primary school or to be streamed into vocational schools and to discontinue schooling upon graduation. Among the youth (15-17 years old) from poor families in Serbia, for example, as many as 23 percent are no longer at school and consider their education completed, and only 4 percent were able to enroll into higher education establishment. The disadvantage can be traced back to the unfavorable learning environment at home and to generally late start of schooling among children from poor families. Publicly subsidized kindergartens provide services to only 30 percent of children, almost exclusively - from non-poor families. Children of the poor are likely to perform worse than pupils from better off families from the start of the schooling, and are screened out of the stream leading to tertiary education or acquisition of more marketable skills. In such an environment, low social mobility should not be a surprise: those who are poor now will remain poor across generations. 54. Public health expenditures in 2001 represented over 7 percent of GDP in SAM (The World Bank [2002c]), and the health financing required transfers from the State budget in addition to health insurance contributions to the health fund. Health indicators are generally better than in countries with similar incomes, reflecting the high levels of education and the 17 universal provision of comprehensive health care services in the past. Maternal mortality rates are low, and they declined between 1990 and 2002. Increases in the mortality rates of children under five years have largely been contained. However, universal access to health care i s being seriously undermined by the inability o f the poor--and some non-poor--to afford the charges now levied by many public institutions--officially or unofficially. As public spending on medical supplies, doctors' wages, and maintenance of clinics becomes tighter, the burden shifts in many cases to patients. This has a clear impact on the poor, as this report documents. 55. The report shows that some emerging problems for poor accessing health and education services have yet to result in a full exclusion of the poor, but they represent a very worrying trend. The analysis of the draft of the PRSP for Serbia and Montenegro shows that most of the problems are well understood by the education and health professionals and policymakers. The PRSP follows a clear set o f priorities developed by the Governments and civil society in SAM. The strategy is based on the premise that an increase in public expenditures generally i s not a viable option given the overall budgetary constraints. Thus, significant efficiency gains should be achieved in each sector. Strategic objectives require some reallocation between sectors. Education i s identified as a priority. The Government of Serbia projects an increase inthe overall spending on education as a share o f public expenditure, while the Government of Montenegro expects to maintain spending on education at its current high level. The report highlights the need for in-depth studies to increase the likely impact of this spending and monitor the impact of PRSP implementation. Reducing Vulnerability:The Safety Net and the Poor 56. In 2001, public safety nets commanded over one-half of the general government expenditures in Serbia and Montenegro (The World Bank [2002c]). Not surprisingly, the effects o f public safety nets on welfare were significant. The household survey data for 2002 allow in- depth analysis of their coverage and efficiency. 57. Social insurance programs (mainly pensions) channeled the bulk of spending on social protection (in Serbia 90 percent of social protection spending), and reached - directly or indirectly - more than half of the population in Serbia and about 45 percent inMontenegro. The 18 main goal of these programs is not poverty reduction but the management of selected social risks. However, these programs have an important poverty alleviation role. While only 16 percent o f combined social insurance spending in Serbia went to the poor (pre-transfer), these transfers covered as many as 61 percent of all poor. For the recipients of social insurance the transfer covered over 70 percent of their consumption, playing a major role as an income source. 58. Social assistance programs, aimed at poor directly, have less resources, and are more numerous and heterogeneous in terms of their objectives, target groups, targeting mechanisms, sources of financing, and implementing agencies. In Serbia in 2002, all social assistance programs combined channeled around 2 percent o f GDP to the beneficiaries. Social assistance benefits reached one in five households, directly or indirectly. In Montenegro, coverage and resources devoted to them were smaller with only 9 percent of population covered by any type o f social assistance. 59. The report shows that for these modest amounts spent on social assistance in 2002, it achieved significant results. In Serbia, various social assistance programs combined covered every third poor person and achieved a reduction in the poverty gap by one -fifth (SLS, 2002). In Montenegro similar analysis of the social transfers is not possible given the data limitations, but the findings of the household survey suggest that the targeted programs (Family Material Support and Reformed Child Allowances) reached almost ten percent o f the poor and transferred significant resources to them. Humanitarian assistance financed by donors significantly complemented Government's programs. 60. The report analyzes an array o f social assistance programs in Serbia and finds that some of them perform particularly well in helping the poor. Figure 5 shows the comparative performance of various programs. From a poverty reduction perspective, more efficient programs would have higher coverage o f the poor, less leakage to the non-poor and greater adequacy. l2 A "perfect" program (never found in reality) would be located in the upper-right quadrant o f Figure 5: it would have 100percent coverage of the poor, and 100percent targeting. l2Coverageof the poor is the share of the poor are covered by the program; targeting i s the share of transfers that are capturedby the poor; and adequacyreflects how important i s the benefit transferredby aprogram inthe consumption of the poor beneficiaries. 19 Figure 5: Coverage, Targeting and Adequacy of MainSocial Assistance Programs in Serbia 100% , , I I I I I I I I I I I I I I 0% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Coverage of the poor: % of poor receiving the benefit Note: The size of the bubbles is proportionalto the relative inponance of the transfer in the consumption ofthe poor Source: World Bank estimations basedon the SLS 2002, see Tesliuk (2003). 61. The targeting performance of Serbia's poverty benefit (MOP) is among the best in the region: as much has 70 percent of all funds goes to the poor (Milanovic inBogicevic et a1 [2003]). For the recipients of MOP it represents a significant help, covering 57 percent of their consumption on average. The program i s thus quite effective. But it has a minuscule coverage (only 4 percent of the poor pre-transfer received MOP in 2002), and the program outlay representedonly 0.1percent of GDP. These errors of exclusion are due to three reasons: (i) very stringent criteria for eligibility (income threshold i s significantly below the poverty line developed for this report); (ii) minimal funding of the program; (iii) insufficient information on the program. As many as athirdof the poor have not heardabout its existence (Milanovic [2003] inBogicevic et a1[2003]); the simulation show that one half among those who would beeligible to receive the assistanceunder the current rules do not apply for the benefit (Mijatovic [2003] in Bogicevic et a1 [2003]). Increased funding, changing eligibility rules, and better informationfor the applicants may bring inmany more poor households to participate inthe program. 62. Figure 5 also demonstrates that there are other programs in Serbia which are less well targeted to the poor. Child allowances, which in 2001 absorbed over 80 percent of social assistance transfers from the budget, are a case in point. According to 2002 (SLS) data they covered just 20 percent of the poor and delivered only 20 percent of funds to the poor 20 benefi~iaries.'~The report recommends that continued effort be made to re-target this program (a process initiated by the Government in 2002). In Montenegro, the public resources for social assistance are almost equally split between the targeted programs (such as Family Material Support and Child Allowances) and non-targeted programs (Caregivers allowances and maternity benefits). The system as a whole covers even a smaller fraction of the poor compared to Serbia. 63. Low coverage of individual programs may not necessarily be a problem. If the social protection system i s built on many programs that address narrow, specialized social risks, then the system as a whole may provide adequate coverage despite the low coverage of the individual components. In S A M social protection system does not come close to this ideal, since all o f the programs combined cover only one-third of the without-benefit poor (in Serbia, even less in Montenegro). Overall coverage o f the system i s significantly lower compared to other countries inthe region (for example, Bulgaria or Romania). Inaddition, there is only a weak coordination between programs. The eligibility criteria for receivingbenefits differ a great deal across various programs (Mijatovic inBogicevic et a1 [2003]). 64. Policymakers in Serbia insist on preserving the full array o f existing programs as they cover very different types of social risks. The funding issue thus appear as a key constraint for improving the targeting: under the current allocation each of the programs i s financed only at the minimum level, and it is really difficult to reallocate the funds between programs without sacrificing an integrity o f those to be cut. To widen the coverage o f the poor while preserving the resource constraint each program needs to be better targeted and the overlap between them reduced. The on-going reform of social assistance in Serbia i s heading in this direction. To ensure the successful outcome of reform it i s important to harmonize the overlapping programs. But increasing the resources available to better targeted programs may be as important. Withdrawal of donor support to the humanitarian programs calls for an increase in the public funding to compensate for the shortfall. Better data, more in-depth analysis and program evaluations can guide further efforts in safety net reform, as the recent comprehensive study producedin Serbia has demonstrated (Mijatovic inBogicevic et a1 [2003]). l3 must I t be noted here that the analysis contained inthe report uses data collected prior to June 2002, when the child allowances program has been reformed inview o f improved targeting. New 2003 data must be used to assess the impact o f this reform. 21 TheRegional Context 65. For this report, a special comparative work was undertaken to produce data on poverty for a number of countries in the region based on exactly the same methodology and definitions. The results, presented in Figure 6, further underline the conclusions on the reality of poverty in S A M . Thus, reducing poverty in S A M will affect positively the situation in the entire region and will provide significant momentum to poverty reduction efforts elsewhere. -eb Figure6: Distributionof Region'sPoor by Countries,Around 2000-02 Montenegro 3% Bosniaand Herzegovina 19% 38% 23% Source: Serbia (2002), Montenegro(2002). BiH (2001), Macedonia (2000), andCroatia (1998)- staffestimatesof absolutepoverty basedon raw data andconsistentmethodology. Kosovo-SLS (2001) from World Bank (2002a). For Slovenia (2000) estimates basedon World Bank - (2001a). Populationdata are for correspondingyears fromWorld Development Indicators (World Bank, 2003) or mostrecentestimates. Strategyfor Poverty Reduction 66. The key finding from the household surveys in Serbia in Montenegro i s very encouraging: the poverty gap i s only around 1percent o f GDP. This figure might suggest that poverty could be dealt with through a direct transfer program. Examining one o f the best performing Serbian social assistance programs (the MOP), one can estimate that with the current efficiency of its targeting it may take 2 percent of GDP (assuming MOP targeting efficiency but ignoring administrative expenses) to eradicate completely absolute material poverty. But such simplistic conclusions are misleading. 67. First, the report shows that poverty in Serbia and Montenegro is a much wider phenomenon than a simple deficit of material resources and lack of adequate consumption of 22 basic food and non-food goods. Equally important are other dimensions of deprivation, such as health and education. Thus, elimination of existing poverty may require more significant resources than estimates basedon poverty gap may suggest. 68. Second, the report highlights the importance of vulnerability to poverty. Many among those who may not be poor at a given point in time in SAM, face risk to slip into poverty due to economic fluctuations or personal shocks. Elimination of vulnerability requires resources that exceedfunds neededfor elimination of absolutepoverty by an order of magnitude. 69. The analysis of the report suggests that even fighting absolute material poverty through social programs is still beyond both the fiscal and the institutional means of the State. Given the very large exclusion errors and the failure to identify the "excluded" poor and their needs, it would be difficult to cover them by an efficient safety net even if more resources are made available to the system. An ongoing reform in Serbia's social assistance system to increase outreach as well as to improve the quality of programs (which would increase the demand for them), will take time to produce results. 70. Several conditions have to be met before social protection system could make a more significant dent in poverty. First, economic growth i s essential to deliver the means to run a more extended safety net. Second, the safety net has to build up a capacity to include all those in need. Moreover, to protect even larger numbers of the vulnerable population (potentially under treat to become poor), a growth-based strategy is the only mechanism for a general improvement in living standards, and of moving large section of the population above the dangerous proximity to the poverty line. However, also for those vulnerable to poverty, the growth based strategy needs to be supported by strengthening the institutional capacity to manage social risks associatedwith transition. 71. For the coming three years it i s expected that Serbia and Montenegro will record a robust growth of GDP of approximately 4 percent per year. This projection depends upon: (i) maintaining macroeconomic stability; (ii)continuation of structural reforms; (iii)public administration reforms; and (vi) regional and international cooperation. 23 72. In particular, the Stabilization and Association process creates an important anchor for policy and boosts its credibility. In addition, significant resources made available as part of this process will help to improve the governance and business environment. The implementation of these policies should lead to the creation of a better business environment that would help to accelerate the growth of the existing private sector, to encourage the start of new private companies, mainly small andmediumenterprises, and to attract more foreign direct investment. 73. These positive changes have to be fueled by redressing the saving-investment imbalance from the over-reliance on external sources, by increased investment, and by ensuring that this goes towards the rehabilitation of infrastructure. The analysis points to the fact that S A M currently has unsustainably high levels of consumption and current account deficit. Public sector deficits have to be brought down by reducing public spending in line with the requirements of fiscal sustainability (see World Bank [2003]). 74. Economic projections underlying the development strategy o f Serbia show that despite GDP growth rate of 4 percent per year, the growth of total consumption in 2004-06 may vary between 1.4 and 1.1 percent per year, and that of private consumption would stay around 2 percent per year. The poverty reduction that can be achieved in the future will directly follow from this consumption growth. 75. Table 4 shows the outlook for poverty reduction. If private consumption increases at the implied rate of around 2 percent per year, with no change in the inequality, poverty will fall to 7.5 percent o f the population by 2006. At this rate, poverty incidence will be reduced by more than a half by 2015, but poverty will not be completely eradicated. 76. Different growth paths offer dramatically different benefits to the poor. Table 4 shows that even slight worsening o f the distribution during the period for projections will undo most o f the growth effects on poverty, and it will fall by 2006 to only around 9 percent from its current level of 10.6 percent. When growth i s combined with more than proportional gains to the poor poverty reduction i s more noticeable (to the headcount index of 6.3 percent by 2006). In discussing the strategic objectives o f poverty reduction and the means to achieve them, it i s worthwhile to separate two related but nevertheless distinct issues: 24 0 How to maximize the growth potential while ensuring its economic sustainability 0 How to maximize the benefits of a given growth rate for the poor i.e., how to ensure its pro-poor character. Table 4: Poverty Projections for Serbia, 2002-06 (percent poor inthe population) 2002 2003 2004 2005 2006 GDPgrowth rate 4.0 4.0 e 4.0 4.0 4.0 Growthrate of private consumption 6.3 3.0 2.0 2.3 2.3 Optimistic poverty projection, pro-poor growth a 10.6 8.9 8.0 6.9 6.3 Baseline poverty projection, equally shared growth 10.6 9.4 8.7 8.1 7.5 Pessimistic poverty projection, pro-richgrowth 10.6 10.0 9.7 9.2 8.9 Notes: aImproving distribution (Gini index falls by one percent per year, consumption of the poor grows by 17 percent against 10 percenton average). Distributionally neutral growth (consumption of an average poor grow as fast as the average, 10 percent over the 2003-06); Worsening income distribution (Gini index increasesby one percent per year, consumption of the poor increases by only 2 percent over the entire period versus 10 percent average); actual; e in 2003, real GDP is expected to grow by about 3.5 percent, with the slight slowdown relative to 2002 reflecting the effectsof drought and a modestfiscal contraction. Source: Staff estimates basedon SLS data and Government Serbia (PRSP 2003). 77. The country needs to generate a substantial GDP growth rate (over the currently projected 4 percent per annum) to allow more sizeable increases in personal consumption and faster reduction of poverty. Or it has to achieve a significant redistribution in favor of the poor - a scenario that i s less likely given limited scope of instruments at the disposal of the Government. Under the current scenario of 4 percent growth of GDP, it meansthat poverty may not change over 2003-06 if the distribution worsens. Moreover, while the aim of reducing absolute material poverty i s an important one, it i s not the only aim of the strategy. Reduction of poverty in other dimensions and reduction of economic vulnerability requires a focus on policies that ensure: (i) stable GDP growth rate, (ii) it translates into the gains to the poor and near that poor at least at the rate of overall growth; and (ii) the poor and near-poor are benefiting more that than proportionately from the public expenditures. 78. The only way to meet these challenges is to consolidate the progress made in reforms to date, to focus on sustainability and stability, and to develop actions to make the growth pro-poor. The broad-based growth will provide sufficient resources to build up the human capital of the poor, strengthen the safety net and address the structural causes of poverty. Therefore, growth has to be accompanied by two types of change to be pro-poor: (i)employment growth, (ii) reallocation of public expenditure towards pro-poor programs. 25 79. These simulations also highlight the importance of accurately monitoring the distributional effects of growth and changes in the poverty rates over time. The implementation of the regular household survey so as to allow accurate monitoring of poverty will therefore be an essential part of the strategy. Datafor Policies 80. Policies based on bad data are misguided, or worse. To avoid the accumulation of problems that are not dealt with, proper monitoring and evaluation must become a routine part of the Government's work. 81. The Poverty Surveys (SLS survey for Serbia and H H S inMontenegro) produced, for the first time in both Republics, a set of reliable and disaggregated poverty indicators. Very importantly, the initiative and leadership for these efforts in Serbia came from the leading PRSP agency - MOSA. The data were analyzed by local researchers and combined with a variety of other data sources. These efforts are fully incorporated in the PRSP. Thus, the first task of creating baseline data on poverty i s successfully completed. 82. The first roundor data collectioninboth Serbia andMontenegro is beingfollowedby the panel survey in Serbia and the follow-up survey in Montenegro in 2003. The analysis of new data and feedback to policymakers will represent the next step in creating national monitoring and evaluation systems for the implementation of PRSP. 83. The following step will consists of institutionalizing poverty monitoring in the regular operations of data collection by State agencies. The poverty data used in this report and in the draft PRSP come from externally funded surveys implemented by NGOs. The ongoing official Household Budget Survey in Serbia i s being reformed with the support of SIDA and Statistics Sweden to play a key role as the PRSP monitoring tool. Starting in late 2003 these data will become available for the analysis and monitoring of poverty. The reform of official survey in Montenegro considerably lags behind. After the completion of the Population Census in November 2003 efforts needto be steppedup to restart the Household Budget Survey. 26 84. For the statistical system to be able to collect, process, analyze, use and disseminate information on poverty and its key correlates, the survey data must be up to this task. In addition, Government action will be required for consolidating the consensus on the official poverty line (in Serbia) or building it (in Montenegro), to establish the poverty measurement methodology that the statistical agencies will follow. Finally, public access to household level data from official surveys will help to strengthen the link between policies and data on poverty. 85. Inthe anaZysis of poverty, several important issues are revealedinthis report as requiring further investigation to improve the understanding of poverty: 0 Chronic versus transient poverty 0 Constraintsto rural incomes 0 Linksbetween poverty andsocial exclusion. 86. Evaluations are the main uses for data to feed in the PRSP. The capacity needs to be strengthened (in Serbia) or built (in Montenegro) to forecast the impact of reforms on the poor, to develop recommendations on instruments that will mitigate the negative social consequences of the reforms, and to improve the targeting and efficiency of the social programs. The report notes pilot experiment on the social impact analysis of electricity tariff reform and the path- breaking ex ante analysis o f options for social safety net reforms conducted by the PRSP working group in Serbia. Such work needs to become a part of the strategy design in all sectors. Mechanisms need to be built to ensure proper feedback from these studies into policies. Some areas require special attention interms of the urgency o f evaluation work: 0 Labor market policies broadly, especially changes inregulations 0 Policies that facilitate the restructuring of enterprises 0 Pension reforms 0 The fiscal and social implications of changes in social contributions, taxes, and benefits geared toward the formalization o f economic activities. 27 87. The Bank and other donors, notably DFID,the EU, SIDA, the Netherlands, Eurostat TA, the EC Food Security Program and the UNDP, are helping to improve statistical collection and develop capacity building training programs in Serbia and Montenegro. The IMF i s providing continuing technical assistance for improvement in statistical methods. The Programmatic Approach to the World Bank's poverty work in S A M helps to ensure that the results of the poverty analysis and monitoring are usedin the process of policy making. Recommendations 88. The report puts forward a number o f general and specific recommendation to support a strategic set o f actions to implement the PRSP in the following areas: (i)macroeconomic policy; (ii) expenditures; (iii)structural policies; (iv) labormarketinterventions; (v)education public and health; and (vi) social protection. These are intended to supplement the efforts of the national Governments inthe implementation o f their Poverty Reduction Strategies. 89. Macroeconomic policies have to focus on reaching overall fiscal and balance of payments sustainability, and the ability to withstand the negative external shocks or exogenous changes o f investor confidence. The goal for supporting the PRSP should be to help to raise domestic saving rates to secure resources for private sector growth, paving the way for productivity gains and employment generation; in the short to medium term this policy has to limit any increases in the overall private and public consumption levels and has to stick to tight targets for the public sector balances. 90. Structural reforms aimed at achieving a pro-poor character of growth need to focus at improving the investment climate for the micro, small and medium enterprises. This offers the best prospects for creating new jobs. Serbia in particular lags behind many countries in the region in the development of SMEs and non-agricultural self-employment. Reforms proposed in the PRSP aimed at improving the business climate will need to be carefully implemented. It will be important to shift the emphasis from special targeted financial incentives for microenterprises to more general policies aimed at improving the rule o f law and level playing field. The on- going Investment Climate Assessment (ICA) study will provide critical insights into the specific measures to assist inthe implementation of PRSP inthis area. 28 91. Labor market policies are strongly emphasized in the SAM PRSP. The donor support should be focused on improvements in the institutional framework. In its present form the employment service lacks the capacity to provide targeted support to specific groups and to facilitate the labor markettransactions. It i s cloggedby large number of claimants who are not in search o fjobs and are seeking the status of unemployed for the social benefits associated with it. It is essential for the success of the labor market reforms to ensure their close coordination with the education sector reform, especially the development of life-long learning. For the successful implementation of PRSP the labor market interventions in a narrow sense should focus on those who are likely to be left from the benefits of economic growth. A particular attention should be given to two groups of workers: the youth, especially those with only basic skills levels, and the pre-retirement age workers affectedby enterprise restructuring. 92. In education, action should be focused primarily on key features of the system that discriminate against the poor: (i) exclusion of the poor from publicly subsidized kindergartens, (ii) shareofthepoorinthegeneral(gymnasiumtype) secondaryschoolprograms, and(iii) low almost complete exclusion of the poor from heavily subsidized tertiary education. The general support o f reform should consist of improving the quality of education for all and the efficiency of resources use. This can be achieved by helping the Government inremoving the existing pro- rich bias in public spending, while providing for the poor better access to educational programs of better quality and relevance. Where there i s a willingness to pay among the non-poor families, more emphasis should be put on fees especially in the pre-school and tertiary levels. At the same time, children from poor families should be helpedby incentives, including financial, to continue their education. It is crucial to ensure successful implementation of the planned special programs to address the handicaps o f children from poor families inthe early years o f schooling. In the secondary school, the general efficiency-improving direction will be towards increased emphasis on broad-based general secondary programs which increase graduate's flexibility on the labor market. 93. In the health care sector the focus should be on improving the governance and finances of the system, and on addressing the issue o f informal payments. The most direct way to improve the financial situation of the system i s to reduce the level of entitlements, which is under 29 consideration in Serbia. The current package i s very generous and includes coverage of treatment abroad and in military hospitals as well as a set of benefits that are non health related such as sick leave. Public procurement of pharmaceuticals in Serbia has historically taken place in a highly controlled marketplace and is ridden by corruption. Despite relatively good health outcomes, SAM i s very much at risk of deterioration of the quality and accessibility of health care and o f a future outbreaks of HIV/AIDS, given existing transmission patterns in the Region. The strategy o f donors will be to support the move to a system, where synergy between public and private providers ensure the population a health care delivery system, which within the framework o f an effective organization and reasonable resources, gives equal access to services based on modem technology and up-to-date scientific methods supported by effective preventive and health promotion efforts. 94. The SLS 2002 data analyzed inthe report provide clear evidence that people in Serbia are paying significantly more out of pocket for health care than official co-payments. Cost of such payments i s deterring the poor, more than the non-poor from using the health care when they need it. This low utilization rate i s especially striking for the rural poor. The co-payment system introduced to formalize widespread informal payments i s rife with exemptions to the point that fewer than 20 percent o f users are required to pay. On the other hand, private health care provision remains rudimentary and serves almost exclusively the non-poor. The direction of reform is to redefine health care providers' roles and responsibilities, tighten eligibility of exemptions, define of a package of basic health care services provided on an equitable basis, and increase cost recovery inthe provision of health services outside the core package. 95. Support to the reform of the social safety net has to focus on assistance to ensure the , successful outcome of on-going reform aimed at harmonizing the overlapping programs and increasing the resources available to better targeted programs. The key challenge for the social protection system is to widen the coverage and reduce the exclusion errors. The Government of Serbia may consider expansion o f the MOP and strengthening the means testing for child allowances. InMontenegro, the resources for the targeted programs (FMS and child allowances) should be increased, while targeting needs to be further improved. Better coordination with NGOs and more pro-active partnerships with the non-profit sector have to be sought to address 30 the challenge of helping the excluded poor. Over the longer term, a thorough assessment of the poverty impact of pension reform options i s needed to facilitate its planning and sequencing. 96. Cross-sectoral programs are being developed to address pockets of poverty and may need a significant support in the PRSPimplementation. Among such programs, regional development programs focus on access to basic services, safety net efficiency, governance and the business environment in the poorest regions. Problems of ethnic minorities and Roma are being included into strategic priorities. The integration of ZDPs and refigees is planned through better government/donor coordination in the cross-sectoral programs aimed at providing long-term solutions. Such programs have to address the livelihoods of displaced populations comprehensively, creating a better environment for employment, solving housing issues and delivering health and education services. Donors may also consider special support to help to mainstream gender inthe process of policymaking. Conclusions 97. For a country with a benchmark of no absolute poverty to now have every tenth person inmaterial poverty -- and every third person with some non-income deprivation -- is a serious concern. On the basis of the nature of this problem, a cross-sectoral approach to poverty reduction i s needed, with a focus on the pro-poor character of growth. There are worrying disparities between the poor and the non-poor in their access to opportunities to develop, protect and use their human capital. If action i s not taken to address these problems, they can impede the prospects for poverty reduction in the future. It is only through sustained pro-poor growth, better targeting and greater efficiency of social protection programs, and an enhanced quality of humancapital investment that SAMwill have the greatesthope of improving living standards for all and of reducing economic risk and vulnerability. 31 32 w M e, 53 33 L O rc e, s -O b L 34 35 36 37 e, I m 38 REFERENCES Atkinson T., Cantillon B., Marlier E. and Nolan B. (2002) Social Indicators: The EU and Social Inclusion, Oxford UniversityPress, Oxford BogiCeviCB., KrstiC G., MijatoviCB.. (2001) "SiromaStvo u Srbiji iReforma Driavne PomoCi SiromaSnima"( The Povertyin SerbiaandReform of the State Assistance to the Poor). Beograd : Centarza liberalno-demokratskestudije Belgrade, 2001 BogiCeviC B., KrstiC G., MijatoviC B., and MilanoviC B.. (2003) SiromaStvo i Reforma Finansijske Podrske SiromaSnima(Poverty and Reformof the Financial Assistance to the Poor). Beograd : Centar za liberalno-demokratske studije : Ministarstvo za socijalna pitania, 2003 (Beograd:Goragraf). EBRD (2000-2), Transition Reports, London. Eurostat (2003), Population and Social Conditions, Poverty and social exclusionin the EUafter Laeken. Reportno No 8/2003. Luxembourg,2003. FederalRepublic of Yugoslavia: Interim PovertyReductionStrategy and Joint Staff Assessment, ReportNo. 24490-YU, July 2002 FSO (Serbia and Montenegro Federal Statistical Office): Statistical Yearbook of Yugoslavia 2000,2001 and2002 2003 Statistical PocketBook Belgrade. Government of the Republic of Serbia (2002) InterimPoverty Reduction Strategy Paper for the Republic of Serbia. Belgrade, June 18,2002. Governmentof the Republic of Serbia (2003) PovertyReductionStrategy Paper InThe Republic of Serbia.FirstDraft. BelgradeApril 24, 2003. Government of the Republic of Montenegro (2002) Interim Poverty Reduction Strategy Paper. Podgorica, June, 2002 IFC (2000). Paths Out of Poverty: The Role of private Enterprise in Developing Countries. Washington, D.C. ISSP-CEPS:MontenegroEconomicTrends, 2000- 2001. ISSP-UNDP (2003). HouseholdSurvey of Roma, Ashkaelia and Egyptians, Refugees and IDPs inMontenegro.Podgorica,2003. 39 KrstiC G. (2003) Profil siromagtvau Srbiji u 2002 (PovertyProfile in Serbia) - in B. BogiCeviC et al. (2003). PoSarac, A. (1997) Unemployment, Labor Market and Social Sector in FRY 1990-1996.World bank, 1997Mimeo. Radevic D. and Beegle K. (2003) Living Standards and Poverty in Montenegro. Background paper for Serbia andMontenegro:PovertyAssessment. ISSPandThe World Bank, June, 2003. Ravallion,M.(1992), PovertyComparisons: a Guideto Concepts andMethods,Fundamentalsof PureandApplied Economics, 56, HanvoodAcademic Publishers Tesliuk, C. (2003) Social Protection and Poverty In Serbia. Background paper for Serbia and Montenegro: PovertyAssessment.The World Bank, mimeo,May, 2003. World Bank (2000a), Making Transition Workfor Everyone: Poverty and Inequality in Europe and CentralAsia, The World Bank,Washington,D.C. World Bank (2000b) Croatia: Economic Vulnerability and Welfare Study. Poverty Reduction andEconomic ManagementSector Unit,Europeand CentralAsiaRegion.Washington,D.C. World Bank (2001a) FRY: Transitional Support Strategy, World Bank, 2001. ReportNo. 22090- YU. World Bank (2001b). Breakingwith the Past: The Path to Stability and Growth. World Bank, 2001. World Bank (2002a) Kosovo: Poverty Assessment. Poverty Reduction and Economic Management Sector Unit, Europe and Central Asia Region. Washington, D.C. Report No. 23390-KOS. World Bank (2002b) Transition: The First Ten Years. Analysis andLessons for Eastern Europe andthe Former Soviet Union, WorldBank, 2002. World Bank (2002~)FRY: Public Expenditure And Institutional Review, World Bank, 2002. ReportNo. 23689-YU. World Bank (2003) SAM: TransitionalSupport StrategyUpdatefor FY 04, World Bank, 2003. 40 GLOSSARYOF TERMS Administrative costs Costs associatedwith the identificationof target groups andthe delivery of programbenefits Adult illiteracy rate The proportionof the population over age fifteen who cannot, with understanding, read and write a simple statement about their everyday life and do simple mathematical calculations. Chronicpoverty A situation where an individualis poor as aresult of long-term structural factors. Consumption(income)smoothing The reduction of fluctuations ina person's (or household's) consumption (income) over time Dependency Ratio Ratio of non-income earning (or dependent) to income-earning members inthe household Employed Anyone who worked at all inthe last seven days preceding the survey and anyone with a permanentjob who has not worked for the following reasons: own illness, maternity leave, a householdmember sick, holidays, education or training, temporary work loadreduction and strike or suspension. Employmentrate Employeddividedby Working Age Population Exclusionerrors Errors intargeting where intendedbeneficiaries are excluded from program benefits Familykhildassistance (allowance) Public cashtransfer basedon the number of children in a household Full-timeemployment Employment i s considered full-time if an individualworks 35+ hours per week. Giniindex Inequality statistic for income or consumption distribution, showing how unequally these are distributedinthe population. It ranges between 0 (perfect equality) and 1(complete inequality). 41 In-kindtransfers Transfers inthe form of goods or services, as opposed to cash Immunizationrate(i.e. against measles) Percentage of children under one year of age who received measles vaccine. A childis considered adequately immunizedagainst measles after receiving one dose of vaccine. Inclusionerrors Errors intargeting arising out of inclusion of unintended(non-poor) beneficiaries inthe program Incomegeneratingprograms Programs designed to generate income that require some form o f contribution (labor, or time, or repayment o f loan). Inthis study it includes public works and credit-based livelihood programs Infantmortalityrate The number of infants, out of every 1,000 babies born in a given year, who die before reaching age 1.The lower the rate, the fewer the infant deaths, and generally the greater the level o f health care available ina country. Informaleconomy The exchange of goods and services not accurately recordedin government figures and accounting. The informal economy, which is generally untaxed, commonly includes goods and services including day care, tutoring, or black market exchanges. Indicator of livingstandards A numerical measure of quality of life ina country. Indicators are usedto illustrate progress of a country inmeeting a range of economic, social, and environmental goals. Humancapital People and their ability to be economically productive. Education, training, and health care can help increase human capital. Laborforce Sum o f employed and unemployed Labor force participationrate (LFP) Labor Force divided by Working Age Population Life expectancyat birth The average number o f years newborn babies can be expected to live based on current health conditions. This indicator reflects environmental conditions in a country, the health of its people, the quality of care they receive when they are sick, andtheir living conditions. Maternalmortality ratio The number of women who die duringpregnancy and childbirth, per 100,000 live births. (Demographic andHealth Surveys andother WHO sources, the United Nations Children's Fund) 42 Meanstest Tests designedto identify and separatethe poor and usually done on the basis of income MillenniumDevelopmentGoals Internationallyagreedgoals for development, derived from the World Summits and conferences of the 199Os,adoptedby 189 nations inthe Millennium Declaration in September 2000. Provide benchmarksfor measuring progress in promoting humandevelopment and poverty reduction untilthe year 2015 and include eight goals: Eradicate extreme poverty andhunger; Achieve universal primary education; Promote gender equality andempower women; Reduce child mortality; Improvematernal health; Combat IV/AIDS,TB, Malaria and other diseases;Ensureenvironmental sustainability; Develop a global partnership for development. Netprimaryenrollment ratio The ratio of the number of children of official school age (as definedby the national education system) who are enrolled in school to the populationof the corresponding official school age. Primaryeducation provides children with basic reading, writing, and mathematics skills along with an elementary understanding of such subjects as history, geography, natural science, social science, art, and music. Out of laborforce (inactive) An individual is considered out of labor force i s she is not employed andnot looking for ajob. Poverty (material) The percentageof the population livingwith consumption below the povertyline Povertygap (depth) The mean shortfallfrom the poverty line (counting the nonpoor as having zero shortfall), expressedas apercentageof the poverty line. This measurereflects the depth of poverty as well as its incidence. Primary healthcare Health services, including family planning, clean water supply, sanitation, immunization, andnutritioneducation, that are designedto be affordable for both the poor people who receive the services and the governments that provide them; the emphasis i s on preventing disease as well as curing it. Privatetransfers Informal transfers made by individuals (households) without government intervention to other individuals (households). Such transfers couldbe incash or kind Publictransfers Transfers madeby the government or its agents to individuals or households Publicworks Government-funded projects to develop or maintain physical infrastructure, largely labor-intensive innature 43 Screening (or targeting) The process of sorting out those inneed of a programfrom those that are not in need Self-selection Targeting through design features that ensure that only the target population makes use of the program, e.g. through setting a wage rate inpublic works that i s lower than market rate - creating a disincentive for those who have employment options inthe market Social assistance A range of benefits incash or kindto provide protection for the most vulnerable persons in society. These programs are usually financed from government revenues Social insurance A range of programs (usually incash) designed to protect individualsinthe event of a decline inincome (due to unemployment, retirement, illness). Benefits are financed through contributions that are usually earnings-related or collected through payroll taxes Transient poverty A situation where an individualis poor becauseof sometemporary shock which couldbe reversed over time Unemployed An individualwho hasnot workedinthe last sevendays preceding the survey, but has lookedfor ajob inthe last four weeks, and i s available to work. Unemployment rate Unemployeddivided by Labor Force Under 5 child mortality rate The probability that anewborn baby will die before reaching age five, if subject to current age-specific mortality rates. The probabilityi s expressed as a rate per 1,000, (UnitedNations Statistics Division's Population and Vital Statistics Report; country statistical offices; Demographic andHealth Surveys and the UnitedNations Children's Fund's (UNICEF) State of the World's Children 2000) Working age population InCEEcountries, population of ages 15to 64 years oldis consideredas working age population. Official statistics defines the working age as 15-54 years for women and 15-59 years for men. Youth literacy rate The percentageof people ages 15-24 who can, with understanding, read andwrite a short, simple statement on their everyday life. 44