WORLD BANK TECHNICAL PAPER NO. 503 Work in progress WTP503 for public discussion May 2001 Household Welfare, the Labor Market, and Social Programs in Albania AlIansoora Rashid Vajeera Dorabawila Richard Adams Recent World Bank Technical Papers No. 414 Salman and Boisson de Chazournes, International Watercottrses: Enhancing Cooperation and Managing Conflict, Proceedings of a World Bank Seminar No. 415 Feitelson and Haddad, Identification of Joint Management Structulresfor Shared Aquifers: A Cooperative Palestinian-Israeli Effort No. 416 Miller and Reidinger, eds., Comprehensive River Basin Development: The Tennessee Valley Autthority No. 417 Rutkowski, Welfare and the Labor Market in Poland: Social Policy during Economic Transition No. 418 Okidegbe and Associates, Agriculture Sector Programs: Son rcebook No. 420 Francis and others, Hard Lessons: Primary Schools, Community, and Social Capital in Nigeria No. 421 Gert Jan Bom, Robert Foster, Ebel Dijkstra, and Marja Tummers, Evaporative Air-Conditioning: Applications for Environmentally Friendly Cooling No. 422 Peter Quaak, Harrie Knoef, and Huber Stassen, Energyfrom Biomass: A Review of Combuistion and Gasifica- tion Technologies No. 423 Energy Sector Unit, Europe and Central Asia Region, World Bank, Non-Payment in the Electricity Sector in Eastern Eutrope and the Former Soviet Union No. 424 Jaffee, ed., Souithern African Agribusiness: Gaining throlugh Regional Collaboration No. 425 Mohan, ed., Bibliography of Publications: Africa Region, 1993-98 No. 426 Rushbrook and Pugh, Solid Waste Landfills in Middle- and Lower-Income Countries: A Technical Guide to Planning, Design, and Operation No. 427 Marinio and Kemper, Institutional Frameworks in Successfiul Water Markets: Brazil, Spain, and Colorado, USA No. 428 C. Mark Blackden and Chitra Bhanu, Gender, Growth, and Poverty Reduiction: Special Program of Assistance for Africa, 1998 Status Report on Poverty in Siib-Saharan Africa No. 429 Gary McMahon, Jose Luis Evia, Alberto Pasc6-Font, and Jose Miguel Sanchez, An Environmental Stuldy of Artisanal, Small, and Mediuim Mining in Bolivia, Chile, and Perul No. 430 Maria Dakolias, Court Performance arouind the World: A Comparative Perspective No. 431 Severin Kodderitzsch, Reforms in Albanian Agricuilture: Assessing a Sector in Transition No. 432 Luiz Gabriel Azevedo, Musa Asad, and Larry D. Simpson, Maziagement of Water Resouirces: Bulk Water Pricing in Brazil No. 433 Malcolm Rowat and Jose Astigarraga, Latin American Insolvency Systems: A Comparative Assessment No. 434 Csaba Csaki and John Nash, eds., Regional and International Trade Policy: Lessonsfor the EU Accession in the Ruiral Sector-World Bank/FAO Workshop, Jlune 20-23, 1998 No. 435 lain Begg, EU Investment Grants Review No. 436 Roy Prosterman and Tim Hanstad, ed., Legal Impediments to Effective Rural Land Relations in Eastern Euirope and Central Asia: A Comparative Perspective No. 437 Csaba Csaki, Michel Dabatisse, and Oskar Honisch, Food and Agricuiltuire in the Czech Republic: From a "Velvet" Transition to the Challenges of EU Accession No 438 George J. Borjas, Economic Research on the Determinants of Immigration: Lessonsfor the Euiropean Union No 439 Mustapha Nabli, Financial Integration, Vuilnerabilities to Crisis, and EU Accession in Five Central Euiropean Countries No. 440 Robert Bruce, loannis Kessides, and Lothar Kneifel, Overcoming Obstacles to Liberalization of the Telecom Sector in Estonia, Poland, the Czech Republic, Slovenia, and Hungary: An Overview of Key Policy Concerns and Potential Initiatives to Facilitate the Transition Process No. 441 Bartlomiej Kaminski, Hiingary: Foreign Trade Issuies in the Context of Accession to the EU No. 442 Bartlomiej Kaminski, The Role of Foreign Direct Investment and Trade Policy in Poland's Accession to the European Union No. 443 Luc Lecuit, John Elder, Christian Hurtado, Francois Rantrua, Kamal Siblini, and Maurizia Tovo, DeMIStifying MIS: Giiidelinesfor Management In,formation Systems in Social Fiinds No. 444 Robert F. Townsend, Agricultutral Incentives in Suib-Saharan Africa: Policy Challenges No. 445 Ian Hill, Forest Management in Nepal: Economics of Ecology No. 446 Gordon Hughes and Magda Lovei, Economic Reform and Environmental Performance in Transition Economies No. 447 R. Maria Saleth and Ariel Dinar, Evaluiating Water Instituttions and Water Sector Performance (List continues on the inside back cover) WORLD BANK TECHNICAL PAPER NO. 503 Household Welfare, the Labor Market, and Social Programs in Albania Mansoora Rashid Vajeera Dorabawila Richard Adams The World Bank Washington, D.C. Copyright © 2001 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing May 2001 1 23.404030201 Technical Papers are published to communicate the results of the Bank's work to the development community with the least possible delay. The typescript of this paper therefore has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. The World Bank encourages dissemination of its work and will normally grant permission promptly. Permission to photocopy items for internal or personal use, for the internal or personal use of specific clients, or for educational classroom use, is granted by the World Bank, provided that the appropriate fee is paid directly to Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, U.S.A., telephone 978-750-8400, fax 978-750-4470. Please contact the Copyright Clearance Center before photocopying items. For permission to reprint individual articles or chapters, please fax your request with complete information to the Republication Department, Copyright Clearance Center, fax 978-750-4470. All other queries on rights and licenses should be addressed to the World Bank at the address above or faxed to 202-522-2422. ISBN: 0-8213-4963-5 ISSN: 0253-7494 Mansoora Rashid is a Senior Human Resources Economist for the Human Development Unit of the Europe and Central Asia Region of the World Bank. Vajeera Dorabawila was a consultant for the Human Development Unit of the Europe and Central Asia Region of the World Bank. Richard Adams is a consultant with the Poverty Reduction and Economic Management Network, Poverty Reduction Unit of the World Bank. Library of Congress Cataloging-in-Publication Data has been applied for. TABLE OF CONTENTS ACKNOWLEDGEMENT .............................................................. v ABSTRACT .............................................................. vi FOREWORD .............................................................. vii I. EXECUTIVE SUMMARY ..1........................................................ A. Introduction ..................................................... 1 B. An Overview: 1990-1996 ..................................................... 1 C. Household Welfare and Poverty, 1996 ..................................................... 2 D. The Labor Market ..................................................... 3 E. Public Services: Developing Human Capital and Poverty Alleviation ..................................................... 4 F. The Challenges Ahead ..................................................... 6 II. HOUSEHOLD WELFARE IN ALBANIA -IN 1996 .............................................................. 8 A. Socio-Economic Characteristics ..................................................... 8 B. Income, Consumption, Amenities, and Assets ........................................... 10 C. A Profile of the Poor ..................................................... 15 mI. THE LABOR MARKET .............................................................. 21 A. The Labor Force ..................................................... 21 B. Labor Market Flexibility ..................................................... 31 C. The Emerging Labor Market ..................................................... 32 IV. PUBLIC PROGRAMS IN RURAL ALBANIA .............................................................. 35 A. Education ..................................................... 35 B. Public Transfers ..................................................... 42 V. REFERENCES AND BIBLIOGRAPHY ............................ .................................. 48 STATISTICAL ANNEX .............................................................. 49 Index of Tables and Figures Tables Table 2. 1: Incidence of Acute and Chronic Illness by Age Group and Location ............................ 9 Table 2. 2: Physical Characteristics of Households .............................................. 13 Table 3. 1: Duration of unemployment for the unemployed ............................................... 25 Table 3. 2: Average Land Area and Value of Land by Region . ............................................. 26 Table 4. 1: Gross Enrollment Rates by Educational Level, Gender and Expenditure Quintile Group ...................................... 38 Table 4. 2: Gross Enrollment Rates by Urban/Rural Sector, Gender and Expenditure Quintile Group ..................................... 39 iii Table 4. 3: Per Capita and Total Expenditure (lek) on Basic, Secondary and Tertiary Education, Albania 1996 ............................................................... 40 Table 4. 4: Ndihme Social Assistance: Households Receiving Benefits ....................................... 44 Table 4. 5: Old Age Pensions: Households Receiving Benefits .................................................... 46 Figures Figure 2. 1: Size Composition of Households by Level of Consumption ...................................... 8 Figure 2. 2: The Incidence of Chronic and Acute Illness by Quintile . .......................................... 10 Figure 2. 3: The Distribution of Consumption ................................................................ 1 Figure 2. 4: Mean Percentage Allocation of Household Consumption Items . .............................. 12 Figure 2. 5: Distribution of Consumption in Urban and Rural Areas ............................................ 12 Figure 2. 6: Ratio of Consumption of Poorest to Richest Quintiles .............................................. 13 Figure 2. 7: Distribution of Household Physical Characteristics by Quintile . .............................. 14 Figure 2. 8: Value of Total Land Owned by Employment Status of Household Head . ................ 14 Figure 2. 9: Incidence of Poverty by Age of Household Head . ..................................................... 15 Figure 2. 10: Distribution of Per Capita Consumption by Region . ............................................... 16 Figure 2. 11: Poverty Rate by Household Size ................................................................ 17 Figure 2. 12: Incidence of Poverty by Number of Children in Household .................................... 18 Figure 2. 13: Incidence of Poverty by the Education Status of the Household Head .................... 18 Figure 2. 14: Incidence of Poverty by the Employment Status of the Household Head . .............. 19 Figure 3. 1: Composition of the Employed, Albania (less Tirana) 1996 ...................................... 23 Figure 3. 2: Characteristics of the Unemployed ............................................................... 23 Figure 3. 3: The Composition of Unemployed By Work Experience ........................................... 24 Figure 3. 4: Characteristics of those in Agriculture-Self Employment ......................................... 27 Figure 3. 5: Characteristics of Wage Employed in the Public and Private Sectors ....................... 28 Figure 3. 6: Wage Distribution in the Public and Private Sectors ................................................. 29 Figure 3. 7: Those Hired Before and After the Election ............................................................... 32 Figure 3. 8: The Characteristics of Workers Hired Before and After the Election . ...................... 33 Figure 3. 9: Median Wages of those Hired Before and After the Election . .................................. 34 Figure 4. 1: Improvements in the Education Status of Successive Cohorts . ................................. 36 Figure 4. 2: Net, Gross and Age Specific Enrollment Rates By Level of Education .................... 38 Figure 4. 3: Public Transfers in Albania as Percent of Total Public Transfers . ............................ 43 iv Acknowledgements The authors would like to acknowledge helpful and comments from World Bank colleagues including: Sue Berryman, Anush Bezhanyan, Christine Jones, Aleksandra Posarac, Jan Rutkowski, and Milan Vodopivec. We would like to particularly thank Harold Alderman for the data set and general encouragement to undertake this task and to Deon Filmer for providing us with helpful suggestions on its use. Most of all we would like to thank Kalpana Mehra for her valuable research assistance to this paper, without which it could not have been completed. We would also like to thank participants of workshops at the World Bank, the Albania Country team and participants of workshop in Albania. The paper was prepared under the guidance of Maureen Lewis, Michal Rutkowski, and Arntruad Hartmann, and has benefited greatly from their comments and suggestions. v Abstract This paper provides an overview of household welfare, labor markets, and social programs in Albania, outside of its capital, in 1996. In that year, Albania was on a cross roads from a period of phenomenal growth to a series of economic crisis, but still ranked as the poorest country in the Central and Eastern European (CEE) Region. The main findings are that the majority of the poor are rural, self-employed in agriculture, a result of Albania's large rural population that is mainly employed in subsistence agriculture. T'hese households also have the highest incidence of poverty, followed by out of labor force individuals and the unemployed. Poverty is highest in the rural north. This is not surprising. At the time of the study the farmers in the north were the poorest, requiring subsidized wheat and cash transfers to survive difficult winters. Interestingly, migration is a major coping strategy in Albania. Household with no migrants were poorer than those where a family member had returned or was currently working abroad. In urban areas, high rate and long duration of unemployment is a key problem, particularly for those with basic and secondary education, for youth and women. The concentration of unemployment among secondary earners, reduces its impact on poverty. The main non-agricultural employer is the public sector; the private sector is quite small. As in other CEE countries, it pays higher and more dispersed wages. Private sector employment is much more flexible than in the public sector-average tenure is short and contracts are verbal. This is consistent with lack of law enforcement and the prevalence of an informal market in the country. Interestingly education has a payoff in Albania; and university educated earn much more than primary school leavers. The study raises concern about the education system and the safety net. It finds considerable drop outs in basic and secondary education among the poor. It finds that education spending is biased against the poor, except in basic education. It also finds that health outcomes (infectious diseases) are particularly worse among the poor-signaling the poor quality of water, sanitation and immunization services in the country. The study notes that outside of pensions, Albania's social protection system (pensions, social assistance) appears is moderately well targeted to the poor. However, high tax rates, and the limited wage base, make a contribution based social protection system questionable. The results show that key challenges ahead for Albania, include fostering the growth of a formal private non-agricultural labor market; ensuring investment in human capital of the poor; moving from social Insurance to safety nets based system for protecting the poor. Finally, the paucity of data on Albania, highlights the need to develop effective information and monitoring systems to inform public policy. vi Foreword Pivotal to World Bank policy on human development and poverty alleviation are investments in health and education, to create a healthy and productive labor force, and safety nets to protect the most vulnerable population groups in a country. This paper provides a first in-depth look at poverty in Albania, outside the capital Tirana. The project grew out of a need to inform the World Bank and Albanian policy makers about the impact of social sector programs aimed at developing human capital and protecting the poor. While the paper is based on 1996 Household Survey Data, we believe that since it represents the only available set of data on poverty and human development, its findings remain relevant for shaping human development policy in Albania. It should also provide a useful benchmark for future studies on household welfare, labor markets and social programs in Albania. Annette Dixon Sector Manager Human Development Unit Europe and Central Asia Region vii 1. EXECUTIVE SUMMARY A. Introduction This paper outlines household welfare in Albania, outside the Tirana region, in 1996. The main thrust of the report is to evaluate household welfare, describe the labor market and assess the equity and efficiency of public expenditures on cash benefits and education. The report has several limitations. First, many changes have occurred in Albania since 1996. The collapse of the pyramid schemes (1997) followed by political turmoil (1998) and the Kosovo crisis (1999) may have changed the profile of household welfare in the country. Second, as the report is based on household survey for Albania outside the Tirana region, it is not nationally representative. While Albania is a largely rural country-with nearly 60 percent of the population residing in rural areas-excluding Tirana eliminates a sizeable population from the analysis. Despite these shortcomings, the report serves a useful purpose in providing a benchmark against which future welfare developments of the population can be measured. While poverty is likely to have increased and deepened since 1996, the basic correlates and determinants of poverty found by this report-agricultural self-employment, unemployment, low levels of education-are unlikely to have significantly changed over time. The report confirms the basic conclusions of the 1996 Poverty Report on Albania' which were based on largely qualitative and some survey data. However, the richer household data available to this report allows a much deeper probe into the causes and determinants of household welfare and the effectiveness of public programs than earlier possible. The paper is organized as follows. The next or second section describes household welfare in Albania and the characteristics of Albanian households; The third section looks at labor market developments in Albania, while the fourth and final chapter evaluate the effectiveness and efficiency of public education and transfer programs. A Statistical Annex is attached to the paper. B. An Overview: 1990-1996 Since 1996, as noted above, a series of repeated crisis have cast a shadow over Albania's transition to a market economy. But, in 1996, the year of analysis for this report, Albania could look back on a four year period of phenomenal growth. During this period output grew by 9 percent per annum in real terms. Major structural reforms, which privatized almost all small state enterprises and all agricultural land, liberalized prices and the trade regime, dismantled many of the controls of the previous regime, accounted for this achievement. 1 Allison et. al, Growing Out of Poverty, The World Bank, 1996 1 The (registered) unemployment rate (collected by the Employment Bureau) fell by half (though this was explained in part by restrictions in eligibility requirements), and inflation rate declined from 240 percent per annum to only 6 percent in end-1995. Employment and real wages increased and the sectoral composition of employment changed. The private sector share of employment grew markedly to nearly 80 percent of total employment. Weak regulatory capacity and high social insurance tax rates also fueled the development of an informal labor market. Finally, as in other Central and Eastern European countries, privatization of agricultural cooperatives and elimination in the restrictions on labor mobility led to a major migration--from rural to urban areas--and from Albania to countries outside the region. Despite progress on these fronts, in 1996, Albania still ranked as one of the poorest countries in the Central and Eastern Europe. The level of GDP per capita--$814--was five times less than that found in the richest CEE countries, such as the Czech Republic, and Hungary, and close to that of its neighbor Macedonia, and the poorer Central Asian Republics. As noted above, the country was, and remains, largely rural2 and agricultural, with agriculture accounting for slightly over half of all GDP. Public services, such as water, roads, sanitation and services were very poor. In 1994, according to the World Bank Poverty Report, Growing Out of Poverty, poverty was pervasive, particularly in rural areas. Nearly 30 percent of the rural population and 15 percent of the urban population were below the officially defined poverty line. C. Household Welfare and Poverty, 1996 In 1996, the rural-urban dimension of poverty was quite similar to that found in the earlier poverty report. The incidence of poverty was higher in rural vs. urban areas and the north vs. the south. This pattern of poverty prevails no matter what poverty line is chosen for the country. The poorest (richest) households in Albania are rural (urban), have a larger (small) family size, more (and few) children. Households with more than three children have the highest poverty rates of all. These characteristics are similar to other Central and Eastern European (CEE) countries. However, Albania is quite different than other countries in the region in some other respects: Unlike other CEE countries, the self-employed in private agriculture are the largest group (60 percent) of poor in the country. In contrast to CEE countries (Romania, Macedonia, Poland), a very small share of the poor are wage earners (9 percent) or the retired (11 percent). However, similar to other countries in the region, the self-employed agricultural workers have the highest incidence of poverty in the country. Households headed by non-labor force participants have the next highest incidence of poverty, 2 In this study, where possible, we compare survey characteristics with those found in other Central and Eastern European countries. However, it is important to note that this comparison is not strictly correct. This is because we are comparing non-Tirana Albania with characteristics of more nationally representative surveys. The rural share of the population was higher than in Bulgaria (33% 1996); Poland ( 55%, 1996) Macedonia ( 45%, 1996); and 52% (1994) Romania. 2 followed by the unemployed. The retired and aged (relative to young individuals) are least likely to be poor in Albania. The education status of the household heads is low relative to CEE norms. In 1996, despite high literacy rates inherited from the socialist regime, and despite considerable progress made in improving the education status of the population over time(see chap. 4), over half of all household heads (average age 50) had basic education or less.3 As in other countries, poor and rural households have the lowest education status of all households in Albania. Living conditions are also poor with adverse implications for health of the population. Despite a high life expectancy at birth-at 72 years-which rivals most countries in the region, Albania has the highest infant mortality rates in Central and Eastern Europe. Public services are very poor for the country as a whole, but the poor are much worse off. They have negligible access to piped water, electricity or sanitation facilities. In contrast, the majority of the richest households have access to these basic services. Not surprisingly, the poor have a higher incidence of infectious vs. chronic disease as compared to richer households. Living with older, more educated, self or wage employed heads in urban areas with a migrant worker (present or absent) in the household significantly reduce an individual's chances of being poor. Living in a households with a young head and in the rural north (relative to urban north) are the two single largest risks of poverty for an individual. These two characteristics are strongly associated with a poor agricultural household. The age, education, wage and self employment of the household head and the presence of migrant worker in the household also matter, but much less so. However, these characteristics have a much stronger impact on reducing the chances of individual poverty in rural areas. D. The Labor Market In 1996, the labor market in Albania was very different than other Central and Eastern European countries. The private agricultural sector-mainly subsistence farming--was the largest employer in the country. The mode of privatization in agriculture resulted in very fragmented and small land sizes. The most well off farmers, located in the coastal plains of the south, own more land, and cultivate wheat crops. marketing some produce. The poorest farmers are in the remote areas of the north, working very small plots and raising livestock. The poorest households require subsidized wheat and cash transfers to survive the winter. 3In contrast, household heads with education level of basic or less are 29 percent in Poland; 22 percent in Romania., but closer to Macedonia (57 percent). 3 The non-agricultural private sector comprised a very small share of employment. The major non-agricultural employer is the public sector. It is difficult to say anything with surety about the private non-agricultural sector because of its small sample size, but as in other CEE countries private wages are higher, and more dispersed relative to the public sector. The incidence of low and high pay is also higher in the private relative to the public sector. Private sector hires less educated individuals relative to the public sector, mainly in manufacturing and construction (vs. health education in the public sector). Employment in the private sector appears flexible-In contrast to the public sector, the average tenure is short and contracts are largely verbal.4 This is consistent with the mass privatization carried out in Albania, the presence of a large informal labor market and the absence of enforcement of labor market regulations. Still, in comparison to other CEE countries, the sector is nascent and undeveloped. Education has a high pay off in Albania. There are incremental marginal returns to education, with university educated individuals realizing the highest increase in wages relative to primary school graduates; followed by secondary school leavers. For the same levels of education and sector of employment, women earn less than men, signaling the potential presence of discrimnination. Private sector workers earn more relative to public sector employees (for each level of education); Industry of employment matters. Workers in arts and culture and communication earn significantly less than other workers. The same results hold in urban areas and in the public sector.5 The rate and duration of unemployment was high relative to CEE countries. In 1996, urban unemployment rates (50%) were far in excess of those calculated from (urban) registered unemployment rates and far higher than in rural areas (11%)6. Unemployment rates were the highest for youth, women, and those with basic and secondary education. The high rate of unemployment was mainly result of its long duration; the flow into the pool of unemployed was very small. The composition of new hires indicates that the most vulnerable unemployed are primary earners, older, women, and those not able to move to rural areas. However, on the positive side, the majority of unemployed are secondary earners, which mitigates their likelihood of being poor. E. Public Services: Developing Human Capital and Poverty Alleviation In 1996, Albania spent over half of its budget (nearly 14 percent of GDP) on social programs. Over half (60%) of social spending was allocated to social protection (pensions, unemployment insurance, social assistance) and the remainder social services (health and education). An important objective of public spending on social programs is 4 There are some caveats. The lack of returns to education, short tenure may be a function of the small number of observations for the private sector. Moreover, in low income countries, verbal contract can be as binding as written agreements. 5 There are not enough observations to estimate separate earning functions for the private sector. 6 Hidden unemployment in rural areas may be one reason for this result 4 to reduce poverty. This study evaluates the effectiveness and efficiency of social protection and education programs. Social Services (Education) In 1996, low levels of public spending on education (relative to other countries) and real declines in education spending levels over the transition, had left social services in disarray. In education, this meant the lack of modem education methods, basic education tools, such as textbooks, and modem teaching aids. In health, infant mortality rates actually worsened over the transition as the provision of basic services declined. Participation in secondary or higher education was low in rural areas and for the poor. Albania has made major gains in education status over time. Younger cohorts are much better educated than earlier generations. This bodes well for the country. However, recent trends of increasing dropouts from the education system are worrisome. A large number of students-mainly poor and rural -drop out of basic education and do not attend secondary school; still more drop out after secondary school. The main reasons are likely to be similar to other countries, such as Macedonia, and include: fees; distance; or costs of school materials (such as textbooks); or (particularly in rural areas) forgone earnings to the individual and his/her family. The high rate of unemployment among secondary school leavers-while likely due to the growth of a less skill intensive private sector-- may also reduce incentives for students to join secondary school. Public Spending on Education was pro-poor only in basic education. Public spending on basic education-which comprises slightly more than a third of total education spending was progressive, and strongly pro-poor. However, as in other CEE countries (e.g. Romania, Macedonia) secondary and tertiary spending are largely targeted to higher income and urban households. Not surprising, the per capita expenditures on education are much higher for the rich than the poor, and for urban relative to rural areas. Social Insurance and the Safety Net. In 1996, the main social protection program, in terms of its share of public spending--was pensions (6% of GDP), followed by social assistance (Ndihme Ekonomike-1.4% of GDP), and unemployment insurance (.8% of GDP). The social insurance programs are financed by payroll tax rates that rank amongst the highest in the region. The study finds that with the exception of pensions, the cash transfer system was pro-poor and reduced inequality. The public pension program The pension system, given its objective as a deferred savings program, is not progressive, or pro-poor. Consistent with the link between benefits and contributions, the pension system tends to increase inequality rather than reduce it. The majority of pension spending is received by the richest households. Very few households in the bottom deciles receive pensions, particularly in urban areas. 5 The unemployment insurance program was pro-poor. The program provides a flat rate cash assistance for a duration of twelve month. By 1996, it was received by a small proportion of the registered unemployed. The program was financed by a tax of 6 percent on payroll, amongst the highest in the region, and its financial status was not well known. However, the program pro-poor and reduced inequality of income (proxied by consumption). The social assistance program was moderately pro-poor. While recent qualitative assessments indicate that social assistance may not be an effective poverty alleviation program, in 1996, The Ndihme Ekonomike moderately well targeted to the poor. There were leakages in the program. Only a third of the poor (bottom two deciles) received benefit. Nevertheless, the program reduced income inequality in both urban and rural areas. F. The Challenges Ahead Recent developments have compromised economic growth, undermined hard won macro- stability, leading to higher rates of inflation and public deficits. These events have also imposed considerable economic and social stress on the population. Poverty, already pervasive, is likely to have increased.7 The recent crisis have also exposed the fragility of a country once thought to be well on its way in its transition from plan to market. The important ingredients for recovery have been discussed elsewhere in Government and World Bank reports. These include political stability in the region in general, and in Albania, in particular, the strengthening of state institutions to ensure the rule of law and improve accountability; financial sector reforms, further privatization of large state enterprises; and agricultural sector reforms to improve titling and registration of land. While policy conclusions on the basis of 1996 data are difficult, and should be interpreted with caution, this report finds four main areas for policy action that would contribute to sustained economic growth and allow a more equitable participation in the growth process. Investment in Human Capital Increasing participation of students in secondary and higher education, particularly for rural areas and the poor, will be a particular challenge to the Government. Improvements in the quality of the education system will also be required to reduce drop-outs and repetition in the education system, particularly among the poor. Fostering the growth of the formal private non-agricultural labor market. Along with investment in education, the expansion in employment and productivity growth in the private non-agricultural sector will be critical for absorbing labor, and raising worker wages, improving the standard of living of workers and their families. Essential to this are the wide range of measures-agricultural, judicial, civil service, financial and land La Cava et. al. (1999), Albania: Filling the Vulnerability Gap, World Bank, Processed. 6 market reforms-listed above. A reduction in tax rates on social insurance (see below) could also help reduce the growing informalization of the labor market. Moving from social insurance to safety nets for protecting the poor As discussed in the social protection strategy for Albania, given the low income status of the country, the primary focus of the social protection system should be poverty alleviation rather than social insurance. The social assistance program also needs to be re-evaluated amid recent growing concerns about its ineffectiveness, and re-considered in the context of an overall social protection strategy which is consistent with Albanian reality. Fostering Policy Evaluation Based on Qualitative and Quantitative Information Developing the private/public capacity to evaluate programs is essential to ensuring that scarce public resources are achieving their policy objectives. 7 II. HOUSEHOLD WELFARE IN ALBANIA -IN 1996 This section first outlines characteristics of a typical household in non-Tirana Albania, and then explores the characteristics of the poor and the correlates of poverty of these households. A. Socio-Economic Characteristics8 Household Size and Composition. The typical Albanian household is headed by an individual about 50 years of age. The size and composition of households differs by geographic region and location (Table A2. 1). The average household has about 6 members, of which, typically two members are children.9 Households in rural 9urban) areas are larger in size, and have more (fewer) children among their members.10 There is some regional variation as well, with household size in the north larger than the south. As in other countries, household size decreases with income. The average household in the lowest quintile has 7 members as compared to an average household Figure 2. 1: Size Composition of Households by Level of Consumption 8 7 34 1 0 1 2 3 4 5 Total Consumption Quintile Source: Albania Living Standard Measurement Survey, 1996. 8 This section reports characteristics of typical Albanian households as demonstrated by the Albania Living Standards and Measurement Survey (LSMS). This was a limited LSMS survey conducted in 1996. The survey excluded Tirana given a household survey of only Tirana was conducted prior to the LSMS survey. As a result the description of a typical household is of those areas other than Tirana in Albania. In contrast, household size is lower in Macedonia (4) and Romania (3). However, household size for these two countries is based on nationally representative data, including the capital city. 10 A child is defined to be 0 to 15 years of age. 8 Size of 4 in the highest quintile. Poorer households also have more children. The average household in the lowest quintile has 3 children as compared to only 2 in the top quintile. Employment A typical household has (less than) two employed members. The number of employed members is much higher in rural areas than in urban centers outside Tirana. In low income households and in rural areas, working members are employed mainly in non wage agriculture, while a wage job is more typical of urban areas and higher income households. The number of unemployed in a household are higher in urban than rural areas, and are higher in the middle (relative to top and bottom quintiles). As a result, the dependency rate, defined as children age 0 to 5 years to workers, is higher for the poor relative to the better-off households. Migration outside Albania is an important source of employment for household members. While only 10 percent of households overall report having migrants, there appears to be a regional concentration of migrants in the urban north. Nearly 15 percent of all households in this region report a migrant, of which, the majority have migrated abroad. Education Despite high literacy rates and high basic enrollment rates, education attainment is not high. About 65 percent of household heads in Albania have achieved only basic education or less (as compared to 50 percent in neighboring Macedonia). Only about 5 percent of the household heads have completed tertiary education. There is substantial regional variation in completed levels of schooling. In urban areas, over 50 percent of household heads have education levels of secondary or higher. In contrast, in rural areas only 25 percent of household heads have secondary or higher education. There are also variations in education achievement by income. In the poorest quintile, Table 2. 1: Incidence of Acute and Chronic Illness by Age Group and Location Acute Chronic Age group 0-5 years 15.2 1.6 6-15 years 3.8 2.5 16-35 years 3.2 3.9 36-64 years 5.4 20.4 64+ years 5.8 49.4 Location Urban 5.26 12.3 Rural 5.65 9.2 Total 5.52 10.3 Source: Albania Living Standard Measurement Survey, 1996. 9 Figure 2. 2: The Incidence of Chronic and Acute Illness by Quintile. 1 8 1 6 EQAcute 1 4 M*Chronic 10) - 8 , 6 2 0 1 2 3 4 5 Total Source: Albania Living Standard Measurement Survey, 1996. about 20 percent of household heads have completed secondary or higher education, as compared to half of all household heads in the richest quintile. Health As in other countries, the young are more prone to acute illness, explaining in part the high mortality rate of children in the country. The incidence of chronic illness increases as the individual ages. There does not appear to be a large difference between rural and urban areas (at least outside of Tirana) in the incidence of acute illness. Hlowever, the incidence of chronic illness is higher in urban areas, even outside Tirana, perhaps reflecting differences in diet, smoking patterns and stress in urban environments. In addition to regional differences, there are differences in health outcomes by level of income. In particular, poorer households suffer relatively more from acute illness- reflecting perhaps poor water, sanitation and limited immunization incidence, while chronic illness appears to be concentrated amongst the better off. B. Income, Consumption, Amenities, and Assets Consumption In 1996, the average household outside the Tirana region reported a per capita monthly total consumption (net of durables and imputed rent) equal to 4,348 Lek ($42.00 per month or $504 per year)." The average monthly level of consumption per capita increases to 5,431 Lek ($52.00 per month, or $624 per year)) if durables and rent are included. The mean level of consumption is higher in urban as compared to rural This report uses per capita consumption is used as a proxy for income. 10 The Allocation of Consumption. Food expenditures account for over half of all household consumption expenditures, and are the single largest expenditure category for Albanian households outside the Tirana region. The second highest category is expenditures on rent. As a result the proportion of consumption allocated for discretionary items is extremely low. Health accounts for 3.4 percent of total expenditures, while private spending on education is almost negligible. Durables and clothing also account for less than 5 percent of total consumption expenditures. The consumption patterns differ in rural and urban areas. Food is a more important consumption item in rural than urban areas, explained in part by the higher per areas. The mean per capita consumption is higher in the south compared to the north.12 . The distribution of consumption is presented in Figure 2.3. The inequality in the distribution of consumption, as measured by the gini coefficient, is 28 percent, a level of inequality that commensurate with that observed in Central and Eastern European countries ((30% Macedonia (1996), 31% in Bulgaria (1997) and .30 in Romania (1994)); and is much lower than that in Russia and Central Asia. For example, the gini coefficient of about 48 percent in Russia (1996) and 54 percent in Kyrgyz Republic (1993). Figure 2. 3: The Distribution of Consumption. 0.3 0.25 Lo Fu~~~~~~~~~~~r A[bania (less Tiran) 0.2 0c 0.15 o 0.1 0 U I\ 0.05- 0 Pf Capita Gc nton p(l Motth (L&k) Source: Albania Living Standard Measurement Survey, 1996. 12 The analysis of this paper uses per capita consumption. However, in the next draft we will provide sensitivity tests to various measures of adult equivalence. 11 Figure 2. 4: Mean Percentage Allocation of Household Consumption Items. Remittances Other Total 0% 15% Durables, transportation & clothing 8% i ~~~~~~~Health 3% Food \ Education 56% 0% Rent - ~~~~~18% Source: Albania Living Standard Measurement Survey, 1996. capita expenditure in urban areas noted above. In both categories of households, out-of- pocket spending on education is negligible, but both types of households spend about the same on health care. Urban areas tend to spend more on rent and durables, while rural households consume more clothing and transportation. As in almost all countries, spending patterns differ by income groups. The mean consumption spending of the poor is roughly a quarter of the mean spending of the rich. The poor spend bulk--70 percent-- of total expenditures, leaving little room for other spending. Expenditures by the poor on health, transportation, durables and education are lower still, comprising less than 10 percent of those of the richest households. Figure 2. 5: Distribution of Consumption in Urban and Rural Areas. Urban Other Rural Other Remittances 9% tm, Rerriat O2t DLurables, 0% trmssponatia. & 0% Transporaion & clothing 1Clotltng 8% 1 W8% W t C Health Ilealth ucation ~~~~~~~0 lq ~ 'I-I _1 5an 0%~~~~~~~~~~~~~0 Rent 53% 26% Food 64't Source: Albania Living Standard Measurement Survey, 1996. 12 Figure 2. 6: Ratio of Consumption of Poorest to Richest Quintiles 35 30- 25- ~20- 15- 10 5 0 Source: Albania Living Standard Measurement Survey, 1996. Living Conditions. The average house outside the Tirana region is built of stone and has approximately 3 rooms, or 1 room/ 2 members. Basic sanitation services are limited. Only half the population has access to piped water and less than half has access to flush latrines. Basic amenities are also limited. Only about 6 percent of the households have a telephone and only 14 percent of the households report having electricity. (Table 2.2) The average number of rooms and household construction are similar across income groups. However, the larger household size in poorer households means more cramped living conditions for the poor. There are marked variations between rich and poor in access to all other household amenities. (Figure 2.7). For example, telephones, piped water, a flush latrine, and electricity appear to be virtually non-existent among the poor. In contrast, nearly 80 percent of rich have piped water, and 60 percent have flush latrines and phones, and nearly all of these households have electricity. Thus, it appears that public utilities are mainly being used by richer households. Table 2. 2: Physical Characteristics of Households Percent of households with Telephone 5.9 Electricity 13.4 Flush latrine 44.3 Piped water 54.5 Stone walls 940 Mean number of rooms in household 2.6 Source: Albania Living Standard Measurement Survey, 1996. 13 Figure 2. 7: Distribution of Household Physical Characteristics by Quintile. 35 --120 E 2 s | ; 80 E 20- . 5 - * | -60 -415 - -n I 44 40 10 = 5- l - 20 a.) a.) a.) ~ .) a. -.o -. -.n -_ - C ~ ~~C= CY C> c:> Telephone 'IMean number of rooms I EMMMMM F-lectricity - Flush latrine *- Piped water -A& Stone walls Source: Albania Living Standard Measurement Survey, 1996. Land Ownership and Value13 Land is an important part of the assets owned by rural and agricultural households. Urban households and wage eamers/unemployed do not hold much of their assets in land . The amount of land owned (Table A2.11), the value of total land owned, and the average value of land/unit of land owned all in general increase with the level of household income . (Table A2.13) . Most of the value attributed to land by rural households comes from crop land. The value of orchards and pasture land is very small. Figure 2. 8: Value of Total Land Owned by Employment Status of Household Head. 600 - ,Y500 - i400 - T 300 - 00- I 00 - 01 Wage Un- Not in Absent Self Agriculture Employed Employed Labor Employed Force Source: Albania Living Standard Measurement Survey, 1996. The value of land is given at the household level. However, there are several employed members in the household and in different occupation categories. Therefore agriculture does not necessarily have to be the head's occupation. As a result, it is difficult to match the occupation with the land owned. 14 C. A Profile of the Poor This section looks more closely at the main correlates of poverty in Albania. It also attempts to identify which characteristics are the most important determinants of poverty status of individuals and households in which they reside. Defining the Poor The poverty line used in this report was defined endogenously from the household survey. According to this definition the per capita income (consumption) of the household with the highest per capita income (consumption) in the bottom 20 percent of the income (consumption) distribution is considered the lower poverty line. This poverty line for Albania in 1996 was 2,888 lek ($28.00 US dollars) per month. This poverty line is slightly less than $1 a day. The measurement of poverty is based on several widely used indicators. The first measure is the headcount index, also referred to as the poverty rate or the incidence of poverty. This measures the proportion of individuals in the population that fall below the officially defined poverty line. By construction, the poverty rate defined by the poverty line is 20 percent of the population is approximately 0.5 million individuals in a population of less than 3 million in 199614. The size of the poverty problem posed by a particular group is measured by the share of the poor in a particular group in the total poor population. In some cases, the poverty rate may be low, but the large share of the group in the total population, may make it a significant share of the poor. The second measure is the expenditure gap index or the poverty gap index which measures the difference between the average consumption Figure 2. 9: Incidence of Poverty by Age of Household Head. 25- ¢ 20- 1 5 Ic 0- __ -a 10 5 0 16-35 years 36-64 years 64+ years Household Head's Age Category Source: Albania Living Standard Measurement Survey, 1996. 14 This is the size of the population in regions other than Tirana. 15 of the poor and the poverty line as a proportion of the poverty line. The greater this measure is the lower the average consumption level of the poor relative to the poverty line and deeper and more serious the conditions of the poor in the country. The final indicator is the gini index which measures inequality in income or consumption. The higher this measure is, greater the inequality in consumption (income). This section also reports results of multivariate analysis of poverty. This analysis allows us to determine which poverty characteristics are the most important determinants of poverty and to evaluate the marginal impact of each characteristic on the probability that an individual is poor. Age The poverty rate falls with age of a household head. The poverty rate of a household head aged 16 to 35 years is almost 10 percent higher than that of a household head aged 64 and above. Similarly, poverty rates for children aged 0-5 are almost double those of individuals of age 64 and above. All else equal, the age of a household head has the single largest positive impact on the likelihood of an individual's being poor. The risk of poverty is almost 30 percent higher for individuals who live with heads aged 16-35 years relative to those living with heads aged 64 or above. The age of the individual also matters but far less than that of the head of household in which he/she resides. Regional Poverty Poverty is a rural phenomenon. The majority of the poor--almost 90 percent-- live in rural areas (Table A2.6); and the poverty rates for rural areas (25%) is more than 5 times the poverty rate in urban centers. This result holds no matter Figure 2. 10: Distribution of Per Capita Consumption by Region. 0.35 0O.3 / -*-Urban 0.25 -Rural 2 0.2 i 0.15 01 K ;0 W_ ME ) cO< s ;°0oz° 0°9°s c°s O°N°c Monthly Per Capita Consumption (Lek) Source: Albania Living Standard Measurement Survey, 1996. 16 which poverty line is chosen for the country. [There is first order dominance of the cumulative distribution of consumption for rural over urban centers]. As shown in Figure 2.9 a much higher proportion of the rural (vs. urban) population has lower levels of consumption. The Gini coefficient for consumption is also much higher in rural areas, indicating a greater inequality in consumption in rural than urban areas. The incidence of poverty is higher in the North as compared to the South. The rural north also has the greatest depth and severity of poverty. Multivariate analysis shows that living in the northern rural household (vs. a northern urban household) has the second largest impact (after age of head) of any characteristic (including education, age of household head, etc.) in increasing the chances of poverty for an individual. However, not all regional differences matter. Specifically, the chances of being poor are about the same if an individual live in a southern urban or rural household relative to a northern urban household. Household Size The poverty rate increases with household size and the number of children in the household. The poverty rate is about the same for households with less than three children, but increases sharply for households with more than three children. (Figure 2.11). In multivariate analysis, household size and the number of young children (those aged 0 to 15 years of age) both increase the risk of poverty, though their marginal impact is small relative to region of residence and age of household head. Gender There is no substantial difference in the incidence of poverty between male headed and female headed households (Table A2.6 and Table A2.7). Multivariate regressions indicate that the impact remains negligible even if we take other household and individual characteristics into account. In contrast in rural areas, a female headed Figure 2. 11: Poverty Rate by Household Size. 60- J 50 50 0 30- 20 0 0L u 10 00 1-4 members 5-7 members 8-10 members 11-16 members Household Size Source: Albania Living Standard Measurement Survey, 1996. 17 Figure 2. 12: Incidence of Poverty by Number of Children in Household. 5 0 40 t320 10'0 0 0 1 2 M ore than 3 Number of Children Source: Albania Living Standard Measurement Survey, 1996. household reduces the chances that a household is poor. It may be that female headed households in rural areas are generally those with a migrant spouse, and as such may receive remittances, reducing their risk of poverty. Education Poverty is strongly linked to education. Illiterate household heads have 30 times the poverty rate as compared to university graduated heads. Household heads with basic education have almost 20 times higher poverty rates compared with university educated household heads. The importance of education is also confirmed by multivariate analysis. An individual living in a household where the head has either basic, secondary or tertiary education is about 13 percent less likely to be poor than a person living with a household head with less than basic education. The importance of education is much stronger in rural areas, where education of a household head reduces the chances of an individual's falling into poverty by as much as 25 percent. The education attainment of an individual also matters, but the education of the household head matters more in reducing his/her chances of being poor. Figure 2. 13: Incidence of Poverty by the Education Status of the Household Head. 30 - 3 25 - Ec20 - 15 - '- _ Less than basic Basic Secondary Tertiary Education Level of Household Head Source: Albania Living Standard Measurement Survey, 1996. 18 Figure 2. 14: Incidence of Poverty by the Employment Status of the Household Head. ;) 35 0 0 v- 30- .o 25 - 20 1 5 5- 0 Employment Status of the Household Head Source: Albania Living Standard Measurement Survey, 1996. Occupation In contrast to most CEE countries, where the majority of the poor are low wage workers and the retired, farmers comprise almost 60 percent of the poor in non- Tirana Albania. The remaining are the retired (11 percent) and the working poor (10 percent), the unemployed and those not in the labor force. The highest poverty rates are for persons with agricultural occupations (this is consistent with other CEE countries, e.g. Romania and Macedonia), while the second highest poverty rates are for household heads that do not participate in the labor force. Household with self-employed heads and wage earners have the lowest incidence of poverty. The incidence of poverty for the retired and unemployed household heads is even lower. Multivariate analysis shows that occupation matters even if we account for all major household characteristics. Individuals residing with wage earner or self employed head of households have a far lower probability of being poor as compared to a household head who is not in the labor force. The risk of being poor for individuals living with household heads who are unemployed, not in the labor force, farmers or absent does not appear to be very different, once other characteristics of heads are taken into account. The occupation of the individual also contributes to the risk of poverty. In particular, the individuals who are self-employed, pensioners or are absent are less likely to be poor no matter the occupation of the household head. Migration Status15 Migration in Albania and the receipt of remittances reflect a household risk management strategy (Table A2.6). The observed poverty rate is zero 15 The sample size reporting migrant status is small and these results should be interpreted with caution. Migration status of the household was determined in two ways using a variable regarding remittances (a question of migration per se was not available in the questionnaire). The question asked if the household receives or did receive remittances from a migrant and asks the sender to be classified as a migrant that has returned home, lives in Albania (one can generally expect this to reflect rural to urban migration) and lives abroad. There were some individuals not included in the household roaster sending remittances. Therefore, 19 among households where the migrant head has returned home; it is slightly higher for households where the individual is currently sending remittances However, poverty rates are the highest in households where there has been no migration. These results are borne out in multivariate analysis. Individuals in households where the head has returned, and was a previous migrant worker has returned, have a much lower likelihood of being poor than one in which there is no migrant. Those households where there are rnigrants currently sending remittances also have lower poverty rates than no-migrant households. (Table A2.6). This result is particularly strong for rural areas, where migration of a household head has a much stronger impact on reducing an individual's risk of poverty. In summary, as in other Central and Eastern European countries, poverty in Albania is largely rural and the incidence of poverty is highest among farmers, less educated household heads, and families with many children. However, unlike other CEE countries, the majority of the poor in non-Tirana Albania, are agricultural households, reflecting the large rural share of the population in the country. Multivariate analysis, which allows us to disentangle the unique impact of each characteristic on the risk of poverty for an individual shows that age (young heads vs. those 64+) and location of a household (rural north vs. urban north) have the single largest positive impact on the risk of poverty facing an individual. The education and occupation of the household head also matter-and an educated household head is far less likely to be poor (relative to less than basic education), as is one that is a wage earner or self-employed (relative to one that is not in the labor force), but the impact of these characteristics is much smaller than that of the age of head and the location of the household. Migration of a worker also reduces the risk of poverty. Household size or gender of household head do not have any significant marginal impact on the risk of poverty. we examine poverty risk by head's migrant status as well as the existence of any member/individual sending remittances to the household. 20 III. THE LABOR MARKET The labor market is the main channel through which economic developments impact household welfare. This section examines the characteristics of the 1996 labor market in Albania in order to gain an understanding of the poverty status of labor force participants described in the previous section.16 A. The Labor Force The non-Tirana Albanian labor market is markedly different than labor markets in the Central and Eastern European (CEE) countries. In 1996, the labor force participation rate (LFPR)17 of the Albanian working age population --68 percent 8-- was higher than most countries in Central and Eastern EuropeI9 and closer to European standards. The non- Tirana Albanian labor force is also largely rural (65 percent) and labor force participants are comparatively younger as compared to other CEE countries. Approximately 36 percent of the non-Tirana labor force is less than 30 years of age. This is high as compared to Poland (25%), Romania (28%); but somewhat closer to Macedonia (30%). The young labor force reflects, in part, the young population demographics of the country. The education level of the non-Tirana labor force also does not compare favorably to other countries in the region. In 1996, while a negligible few in the labor force had less than basic education; the proportion of the labor force with only basic education or less was much higher than in Macedonia (40 percent); Poland (25%); and Romania (12%). A 16 While the section, like the report, covers the labor market in non-Tirana Albania, and has obvious limitations, some basic parameters of the labor market such as the labor force participation rate; employment rate; private/public share of employment and relative wages in public/private sector obtained from the 1996 non-Tirana survey are broadly consistent with the employment data published by the Statistical Office and included in recent World Bank reports. For example, Tang et. al., Country Economic Memorandum: Albania, World Bank, 1998. 17 The labor force participation rate is 73 percent (1995) This reflects a major decline in labor force participation rates, from about 84 percent in 1989. As in other CEE countries, the decline in labor force participation rates was much more marked for women, as compared to men. This reflects an alignment of the very high pre-transition labor force participation rate for men, and particularly for women towards OECD norms. (Albania: Growing Out of Poverty, World Bank, 1996). The comparison of the labor data to other countries with nationally representative data should be interpreted with caution. The labor market data on Albania excludes Tirana and therefore is not nationally representative and has a slightly rural bias. 18 Those over 60 years of age are excluded from the total at this point. 19 Labor Force participation rates (LFPR)are 55 percent in Macedonia; 52 percent in Bulgaria; 53 percent in Romania; and 61 percent Poland. (Poverty Assessment Reports; Rutkowski, Welfare and the Labor Market in Poland, World Bank, 1997). 21 120 Lk te E small proportion of the labor force-6 percent-- has tertiary education. Like other CEE countries, women form a smaller share of the labor force as compared to males. A significant share of the working age population, remained out of the labor force. Those not in the labor force include primarily housewives (38%) and the retired (41%), but also students (12%) and discouraged workers (9%). Working age adults outside the labor force are about evenly divided between urban and rural areas and have slightly lower levels of education as compared to labor force participants. Non-participants are either old or young. In contrast to the labor force, few are between 30-50 years of age. A small proportion-6 percent--of the working age population was absent from their usual place of residence. This likely reflects migration to Greece, Italy or other countries as well as to urban areas in search of employment. Given the widespread migration in search of employment, and aversion to reporting income from abroad, the actual figures of those absent may actually be higher than reported.21 The majority of these are male and from the rural south. Absent individuals are generally younger than labor force participants-about three-quarters are less than 40 years old-and have education levels typical of the labor force. The Unemployed T'he survey based unemployment rate in non-Tirana Albania was approximately 24 percent in 1996. The unemployment rate in Albania was higher in comparison to other countries in the region where survey unemployment rates are available (e.g. Poland, Romania, Bulgaria, Russia), but far lower than the unemployment rate found in neighboring Macedonia (32%) and Yugoslavia (39%). The urban unemployment rate was very high--nearly 50 percent --and much lower--1l percent--in rural areas. Unemployment rates were also higher in the urban south relative to the urban north. This means that unemployment is a much more severe problem in urban areas than reflected in the 12 percent (urban) registered unemployment rate. One reason for low rates of rural unemployment may be the presence of hidden unemployment in this region, but low rural unemployment may also be the result growth in private sector job opportunities in agriculture. As in OECD countries, unemployment rates are highest among youth and women. University graduates have relatively lower unemployment rates compared to individuals with secondary or basic education.22 20 The proportion of the labor force with tertiary education is 8 percent (Romania); 9 percent (Poland); and 14 percent (Macedonia). 21 Some migrants, depending on the length of absence, could have been completely omitted from the household roaster. Unemployment rates are higher for the middle income households than the poorest or richest quintile. 22 Figure 3. 1: Composition of the Employed, Albania (less Tirana) 1996. G overnment WaState Wage 8% Private Wage 6 % Self employed 7% Agriculture --- 72% Source: Albania Living Standard Measurement Survey, 1996. The majority of unemployed live in the urban south. The typical unemployed worker mirrors the characteristics of groups with high rates of unemployment. He/she is is young - almost three quarters of the unemployed are aged 16 to 35 years of age (Figure 3.2); and about half are less than thirty years old. The unemployed are roughly evenly divided into those that have basic education or have completed secondary school.23 A negligible share of the unemployed have university education. As in most OECD and Central and Eastern European countries, an overwhelming majority of the unemployed have previous work experience. Only 12 percent of the unemployed are individuals who have never worked and as such, are new entrants to the labor force. Almost all new entrants are less than 24 years of age and are evenly Figure 3. 2: Characteristics of the Unemployed 80 70 60 50 40 30 20 1 0 0- S S 2~2 Figure 3. 3: The Composition of Unemployed By Work Experience Never- employed 127to Short-term 16% o n g-ter m~~~Lg-er 72% Source: Albania Living Standard Measurement Survey, 1996. distributed between rural and urban areas. The majority--about 60%--have completed secondary or higher education. In contrast to neighboring Macedonia, where the majority cf the unemployed are new entrants to the unemployment pool, exit is a more important reason for being unemployed relative to restrictions of entry. The flow into unemployment-as measured by the share of unemployed who have been in this status for 3 months or less-is quite low. Thus, as in Macedonia, the high unemployment is largely the result of the long duration of unemployment. The duration of unemployment is long in Albania.24 Nearly 80 percent of the unemployed have been in this status for more than 12 months; and about 60 percent have been unemployed for three or more years. The duration of unemployment is longer than in OECD and most Central and Eastern European countries. For example, in Bulgaria the incidence of long term. unemployment is 60 percent while in Poland 40 percent of the unemployed have been looking for jobs for the last 15 months. Only neighboring Macedonia has a comparable rate of long term unemployment (80 percent). The job search duration for the unemployed is also quite long. The average unemployed person has been searching for a job for nearly 3 years. The majority of those with long duration of unemployment are also of prime working age. The long term unemployed have slightly lower levels of education and are more likely to be women as compared to individuals who have never worked or are short term unemployed (have been unemployed for less than a year); but, like short term unemployed, are predominantly urban residents. 26 The long duration of unemployment does not bode well for the chances of the unemployed to find a job. Long term unemployment often depletes market skills and 24 Long-term unemployed are those currently unemployed who have been in this status for more than 12 months. 25 The comparison is with nationally representative survey data. (Rutkowski, 1998). 26 It is only somewhat higher for the 25-29 year olds. 24 Table 3. 1: Duration of unemployment for the unemployed. Duration of Percent Average Length of Unemployment Unemployed Unemployment (months) I Monthor less 1.83 1.00 1-5 Months 5.53 3.17 6-12 Months 13.62 10.14 1-2 Years 5.92 17.71 2-3 Years 13.08 25.53 More than 3 Years 60.03 54.36 Total 100 38.59 Source: Albania Living Standard Measurement Survey, 1996. signals less productive workers to employers.27 On the positive side, unemployment appears to affect secondary workers. Nearly 60 percent of the unemployed live in households with at least one employed member. This may explain the low poverty rate found among these individuals (relative to those not in the labor force or agricultural households). The Employed In 1996, nearly 80 percent of the labor force was employed.28 The employment rate of the population-53 percent--is higher than regional norms (Macedonia (38%); Bulgaria (45%); Romania (48%); Poland (less than 52%)). Self-employment in private agriculture was the main source of employment. Nearly 72 percent of the employed were engaged in farming activities. This aspect of the labor market in Albania is strikingly different than other Central and Eastern European countries and much closer to low income countries. The share of wage employment is very low--only 23 percent of all employment. The majority of wage earners (60 percent) worked in the public sector, in state enterprises and the civil service; The remainder-a very small share-of total employed worked for wages in the private sector. An even smaller share of total employment was comprised of self-employment in the private non-agricultural sector. Thus, outside the Tirana region, the private non-farm sector formed an almost negligible share of total employment. 27 As in other countries, those indicating they are unemployed may be engaged in informal market activities. However the survey does not preclude informal market activity as private sector work. Indeed, the presence of many 'non-contractual' jobs may be an indication that the private sector is represented in the survey. The other option, found in Bulgaria, may be that the unemployed have reservation wages that are too high relative to their realistic prospects in the labor market. 28 In 1996 official statistics reported an employment rate of 76 percent and an unemployment rate of 11 percent; with 13 percent of the labor force estimated to be employed abroad. However the composition of employment is largely consistent with official statistics: 50 percent in private non-agriculture; 16 percent in the public sector; 10 percent in the non-agricultural private sector. 25 Agricultural self employment In 1996, as now, agriculture contributed nearly 60 percent of total GDP. Individuals employed in private agnrculture were young, with over half of those engaged in agricultural self-employment between 16 and 35 years of age, and about equally divided into males and females. The average level of education was also quite low relative to the average labor force participant. An overwhelming majority-or 80 percent-- of the work f orce had only achieved basic education or less. (Figure 3.6). Despite large productivity gains and output gains realized by the private agricultural sector, the majority of agnrculture, as in other low income rural economies, is based on subsistence farming.2930 The mode of pnrvatization of cooperative land, based on equity of land size and value, has resulted in an average land holding that is very small--only 1.3 ha per holding--and a high degree of land fragmentation. The slow pace of land registration and the ensuing thin land market has also contributed to small land holding size. The large growth in agricultural productivity realized after the privatization of cooperatives, resulting from a combination of increased yields and a shifting in land use patterns, was mainly used for increased home consumption. As in other mainly agricultural low income countries, a very small share---approximately 10 percent-- of gross farm income is marketed, with most of the marketed output concentrated in higher income farms. As a result, poverty rates in self-employment in agriculture are the highest in the country; and the self-employed in private agriculture comprise the bulk of the poor. There was a large variation in the agricultural production and farm income in Albania. Agricultural activities are concentrated in three main geographical areas: (i) the coastal plains; (ii) the upland areas; and (iii) the mountains. The most fertile areas, Table 3. 2: Average Land Area and Value of Land by Region. Rural North Rural South Average area of land 4.02 8.41 Number of households 574 410 Average value of land ' 55,153 97,188 Number of households 574 410 Average value of crop land l 55,767 99,038 Number of households 545 397 'The average area of land is calculated based on those that report owning, using state owned land or renting land for agricultural activity. It is calculated per household and the unit of measurement is given in dynyms. 2 The average value of land is calculated for those that report owning land and is calculated per household in lek per dynym. Source: Albania Living Standard Measurement Survey, 1996. 2) The background in this section draws heavily from the Albania Poverty Report, mainly because the basic characteristics of the self employed in private agriculture are not likely to be much different in 1996, (despite greater migration and further improvements in agricultural productivity or growth) the year of this survey and 1994, the year of analysis of the Poverty Report. 30 Albania, Growing Out of Poverty, World Bank, 1997. 26 Figure 3. 4: Characteristics of those in Agriculture-Self Employment I I 90 60 40 30 20 10 ~ ~ ~ ~~4 Source: Albania Living Standard Measurement Survey, 1996. comprising the most valuable land, are in the coastal plains of the South. Compared to other regions, the land size of coastal farms is larger, and farms tend to be wealthier (as measured by gross farm income and value of land owned). Agricultural production is mainly focused on crop farming, comprising wheat and fodder (maize and alfalfa). Livestock production is important as well, but comprises a small share of agricultural production. The region also markets larger share of its wheat production relative to other regions. In contrast, in the mountainous and upland region of the North, land sizes are small, and average gross farm income and average land values are lower. Wheat production is important share of production but the predominant agricultural activity is livestock. Most of the agricultural output is consumed at home, and very little is marketed. This diversity of agricultural production between the north and south contributes to the high incidence of poverty in rural areas relative to urban areas, and the higher poverty rates found particularly in the (rural) north relative to the (rural) south. Agricultural households rely on diverse incomes to survive. For the average rural family, wages are an important source of income. For the poor, earned income exceeds farm income. The poor also rely on low rural pensions to survive. The poorest farms do not have enough land to produce for the household, and are also reliant on social assistance and subsidized wheat/flour provided by the state in the winter months. The large share of employment in agriculture likely reflects under-employment in rural areas. In 1994, the Poverty Report on Albania estimated that nearly 24 percent rural population (almost 500K people) were surplus to carrying capacity of the agricultural land. The importance of wage income, and a returned migrant for reducing the risk of poverty in rural areas, noted in the previous chapter, and the large under employment in 27 private agriculture has contributed to the large migration of Albanians from rural to urban areas and from Albania to other countries. Wage employed The three main sources of wage employment in Albania are: (i) the private sector; (ii) the state-large state enterprises have been difficult to privatize and continue to employ workers; and (iii) the Government (civil service). Despite changes in the composition of employment described above, the public sector (comprising the Government and state enterprises) remains the major employer in the country. In 1996, outside the Tirana region, public sector jobs comprised over two thirds of all wage jobs in the country, with the Government and state sectors holding about an equal share of these jobs. The remainder, or only one-third, of all jobs were in the private non-agricultural sector. The highest concentration of wage jobs are in manufacturing (22%) and construction (10%), followed by science and education and health care (23%). Nearly 60 percent of all jobs in the Government sector are in the areas of health, education, sciences, army and police. Employment in the state sector is mainly oriented towards manufacturing, followed by science/education and communication. In contrast, private employment is concentrated in manufacturing and construction (55% of all private sector jobs), and in trade, commercial services and other productive activities (21%). The typical person engaged in wage employment was male and resided in the urban South (Figure 3.5). Wage earners tended to be more educated than agricultural self employed. Nearly 2 in 3 people engaged in wage employment (vs. 1 out of 5 in the private agriculture) report at least a secondary level of education. Wage earners are also on average older than those engaged in private agriculture. An overwhelming majority (or Figure 3. 5: Characteristics of Wage Employed in the Public and Private Sectors 80 70 01 Public 60 * Prvt 50 40 30 I20 0 4~ 4 Source: Albania Living Standard Measurement Survey, 1996. 28 over 60 percent) of those in wage employment (vs. 40 percent in agricultture) are aged 36 to 64 years of age. The small size of the private sector makes it difficult to say anything with surety about this sector. However, with this caveat in mind, we note some emerging differences in the characteristics of wage employed in the private and public sector. In the public sector, a typical wage earner is male, between 35-64 years of age, with at least secondary levels of education or higher (Figure 3.6). In contrast, private sector wage earners are younger, aged 16 to 35 years of age, and are far less educated. Nearly 95 percent of private sector workers have secondary or lower education, as compared to only 80 percent of public sector workers. There are some similarities across both sectors. Most wage jobs, whether public or private, are located in the Southern region and in urban areas. Both sectors also employ more males than females, though the gender gap in wage employment is larger than for the private sector. The level and distribution of wages. The average wage in Albania was approximately $73 in 1996. There was marked variation in the wages across private and public sectors.31 The average wage (exclusive of allowances) was the highest in the private sector ($90), followed by mean wages in the state sector ($72). The average wage Figure 3. 6: Wage Distribution in the Public and Private Sectors 0.6 0.5 Private 0.4 , \ - - - ~~~~~~~~~~Public Monthly Wage (Lek) Source: Albania Living Standard Measurement Survey, 1996. 31 According to the Poverty Report, the average public sector wage in Albania was $83 in for the first half of 1996. 29 received by Government employees ($62) was the lowest wage of all three sectors.32 The inequality in the wage distribution, as measured by the Gini ratio for wages was .30, higher than Macedonia, but lower than in Bulgaria and Poland. The inequality of wage was higher in the private sector (Gini of .42) relative to the the public sector, including the Government (.25) and the state (.23)33 As shown in Figure 4.6, the private sector has a higher incidence of low and high paid jobs relative to the public sector. The determinants of the level and distribution of wages. What factors explain differences in the level and variation in wages across individuals? The main determinant of wages in Albania is the level of education.34 Individuals with university education earn incrementally more than secondary school leavers; and secondary school graduates earn more than primary school leavers. Gender is also very significant. For the same level education, experience, sector and location, women earn nearly 24 percent less than men. This indicates the presence of discrimination against women in the labor market. Earnings also differ significantly by industry of employment. As in other formally socialist countries, work experience, does not explain much of the difference in wages across individuals. The experience gained in socialist period has become irrelevant over time. Employment in the private sector also raises earnings significantly, even controlling for other characteristics. These results are mirrored for the public sector and for urban areas.35 I'he main source of wage inequality is the industry of employment. This factor explains 20 percent of the total variance in wages. Industry of employment also accounts for a large proportion of the explained variance in wages (Table A3.16). The second most important factor is education which explains 16 percent of total explained variance, followed by gender (11%) and private sector employment (10%). This result is quite different than for CEE countries (other than Bulgaria) where in general education is the main determinant of wage inequality. The characteristics of low paid workers. The poorest workers are those with low wages. As can be seen in Figure 3.6 above, the incidence of low paid employment is highest in the private vs. public sector (25% of all jobs are low paid jobs vs. 19% for 32 The low wages received by Government relative to the private sector was also reported by the CEM (1997). 3 The dispersion of wages as measured by the wage decile ratio was 3 in non-Tirana Albania. Wages were more dispersed in the private sector (5.3)33 as opposed to the Government (4.2) and the state (3.0) The decile ratio for the Government obtained by the survey is broadly consistent with that reported in the 1997 CEM (3.7). 34 The explanatory power of the regression, as measured by R2 is (.19), and is comparable (though somewhat lower) than estimates of similar earning functions for Macedonia (0.23); Poland (0.34); Romania (0.29); and Bulgaria (0.27) 5 The very small number of observations for the private sector make it difficult to estimate robust earning functions for this sector. 30 Government), and for those with basic education or less. However, the private sector also has a higher incidence of highly paid workers. The incidence of low pay in roughly the same in the north and south, as well as between rural and urban areas. However, as a result of high population shares, low paid workers are concentrated in the rural south The incidence of low paid employment is much higher among females than males (Table A3.15). However, given the large share of men in employment, low wage employment is a larger problem for men than for women. Low wage employment is concentrated among the young, probably due to life cycle effects (Table A3.15). Most low paid jobs are located in manufacturing and science and education. B. Labor Market Flexibility Labor Contracts. The flexibility of the labor market is determined in part by the type of contracting arrangements prevailing in the country. In 1996, three types of contracting arrangements between employers and employees were prevalent in Albania: (i) a termless contract-providing employment without any time duration; (ii) fixed term contract- providing employment for a fixed time period; (iii) verbal (no written) contract-work is carried out under a verbal rather than written commitment. The predominant contracting arrangement-in 60 percent of all job contracts-was a termless contract, followed by no contract (20%) and fixed term (11%). Not surprisingly, the majority of public sector jobs were under a termless contract. In contrast, most private sector contractual relationships (55%) were not formalized, but were verbal in nature. The majority of contracts in construction, trade and commercial services sectors have no formal contract or are fixed term contract. A very small share of private sector contracts were termless. This evidence suggests that the private sector in Albania is quite flexible relative to the public sector. Length of Job Tenure Another indicator of labor market flexibility is the length of job tenure. Recent studies have suggested that longer average tenures are associated with more inflexible labor market structures, such as higher severance pay, etc.). For example, in the US, considered a flexible labor market, the average length of tenure is about 7 years, while in Germany, generally considered a more restrictive labor market, it is about 10 years. In Macedonia, one of the more restrictive labor markets in Central and Eastern 36 Europe, the average tenure duration is 14 years. The average length of tenure in a job in Albania-9 years--is close to the restrictive OECD level. The length of tenure varies by private and public sector. In the private sector, the average tenure is only 2 years, while it is almost 13 years in the public sector. Public sector jobs such as science, education, health care have the longest average tenures. In comparison, average tenures in construction and trade are very low. This evidence, in addition to a wider variation in wages in the private sector, would indicate that the private sector is quite flexible relative to the public sector in Albania. There are some caveats. First, in low income countries, with tight community networks, 36 Data on other countries from note prepared by Rutkowski (1999). 31 verbal contracts can be as binding as written contracts. Second, the large downsizing of the state that has already taken place, coupled with limited legal enforcement systems, implies that formnal contracts may also not be binding in the public sector. Third, while low tenure in the private sector may reflect the small size or even the relatively recent emergence of the sector. C. The Emerging Labor Market As noted earlier, the labor market characteristics described for 1996 above were the result of major shifts in the labor market. How did the wage and employment characteristics of workers change over time in Albania? The changes in the labor market that occurred as a result of the transition to a market economy also provide some insights into labor market flexibility. The household survey used to prepare this report does not provide us with a time dimension. We therefore assess changes in the labor market structure by assessing the wage and socio-economic characteristics of individuals hired before and after the 1993 election.37 In the early transition years, the majority of jobs were concentrated in the state sector (cooperatives and other state enterprises); the Government was the next largest employer. This situation changed dramatically over time. Employment in the Government and the state sector contracted, while the private sector increased its share of employment.38 The survey data therefore confirm these national statistics. In contrast with pre-election years, new hires are now mainly concentrated in the private sector. New hires Figure 3. 7: Those Hired Before and After the Election. Before Election After Election Government Private 21% 7% Government Pn va e t S taa 58% 2% Source: Albania Living Standard Measurement Survey, 1996. are male, younger (half between 15-34 years of age) and less educated than individuals hired prior to the election. Most of the new hires are rural, but the urban bias has decreased after the election. In both public and private sectors, new hires are 37 It is important to note that the employment situation prevailing for those hired post 1993 is not the all 1996 situation; as the latter would be based on all workers, both those hired before and after 1993. 38 About 60 percent of those hired within the last 3 years were hired by the private sector (Table b.7). 32 predominantly male (vs. a more equal distribution of sexes among those hired prior to 1993). There are differences in the characteristics of new hires between public and private sectors. Post election, the public sector has hired mainly secondary and tertiary graduates for jobs in science/education, health, army/police (Government) and in manufacturing and construction (state). Public sector hiring has a slight rural and southern bias. In contrast, new hires in the private sector are less educated and mainly female. Over half have basic education or less (vs. less than 20 percent in the Government); and only 6 percent have tertiary level education. (vs. 32% in Government). The private sector new hires have a similar age distribution to Government and are predominantly in low paying manufacturing and construction jobs The transition also changed the level and distribution of wages over time and across sectors. Prior to 1993, the state sector paid higher median wages than either the Government or the private sector. After 1993, the entire situation changed. The private sector began to wages that were higher on average than the Government or the state. The wage distribution in the private and public sector before and after election is difficult to construct because of limited data. However, as in other countries in the region (Romania, Macedonia, for example), the emergence of a private sector with a wider dispersion of wages than the public sector is likely to have caused overall wages in Albania to have become more dispersed over time. Figure 3. 8: The Characteristics of Workers Hired Before and After the Election. 70- 60 - N~~~ Hired before election 7 Hired after election 50 i 40 - 30~ ~ ~ ~ ~~~~3 Figure 3. 9: Median Wages of those Hired Before and After the Election. 8000 |Hired Before Election |-Hired After Election 7000 -_-_ 6000 3 4000 =:3000 2000 I 00 0 0 Government State Private Total Source: Albania Living Standard Measurement Survey, 1996. In summary, in 1996, the Albanian labor market was markedly different from Central and Eastern European labor markets, and resembled labor markets found in low income countries.39 The majority of private sector employment was concentrated in agriculture. The non-agriculture sector rewarded education but was dominated by the public sector. The characteristics of new hires in the emerging labor market matched the skill composition of unemployed, but did not favor female or urban residents (unless they are willing to move to rural jobs). The high rate and long duration of unemployed is also worrisome as it depletes worker skills and reduces his/her chances of finding a job. The growth of the private non- agricultural employment will be critical for employing not only the mainly young urban unemployed, but also for absorbing the continuing stream of migrants flowing from rural tc, urban areas as agriculture continues to shed surplus labor, and as a natural course of economic development, the employment in the agricultural sector declines. Facilitating the growth of a formal private labor market that rewards skills will be critical for raising labor productivity and spurring' economic growth. While it is difficult to say anything with surety about the private non-agricultural labor market because of its small size (in the sample), the private non-agricultural labor market appeared to have some facets of private labor markets in other CEE countries. The private sector paid higher wages and had more dispersed wages (vs. the public sector). Consistent with the limited legal enforcement systems in the country, and a large informal sector, the private sector appears more flexible than the public sector in terms of short job tenures and the prevalence of verbal contracting. There are some caveats. The recent mass privatization indicates that the public sector may not be inflexible. Moreover, verbal contracts are not necessarily less binding, and the short length of tenure may reflect the recent emergence and small size of the sector. As noted, the survey data do not include Tirana, as such the comparison is not really valid. However, the cDmposition of employment in the non-Tirana survey data, though it likely underestimates civil service and non-agriculture private sector employment, is quite similar to national employment distribution. .34 IV. PUBLIC PROGRAMS IN RURAL ALBANIA In 1996, Albania spent over half of its budget and nearly 14 percent of GDP on social programs. Over half of total social expenditures were allocated to social protection (pensions, unemployment insurance an social assistance) and the remainder on social services (health and education). Education and pubic transfers are two Government programs designed to build and protect human capital. Education represents an important means of improving the stock of human capital for future generations, and public transfers can be used to protect the livelihoods of those who are left-behind during the transition to a market economy.40 This chapter will examine the impact of the education system on human capital, equity and poverty alleviation in Albania. It will also evaluate the efficiency and effectiveness of the public transfer program. A. Education Education is crucial to the social and economic development of Albania, where more than 40 percent of the population is of school age. This section discusses the educational system in Albania by focusing on the three general levels of education: (1) basic schooling (ages 6 to 13); (2) secondary (general and vocational) (ages 14 to 17); and tertiary (university and vocational) (ages 18 to 24). Albania entered the transition phase to a market economy with high levels of participation at all levels of education. As shown in Table A4.1, in 1990 gross enrollment rates were above 100 percent at the basic and lower secondary levels, and almost 80 percent for upper secondary. Comparison with other countries in the region suggests that, with the exception of tertiary education, these enrollment rates were generally above average. Because of such high enrollment rates, illiteracy is virtually non-existent in Albania. According to Figure 4.1, the illiteracy rate among individuals age 14 to 64 is 3 percent or less.41 Moreover, the education status of the population has increased over time. While current household heads have lower education status as compared to CEE countries (see Chapter one), about 45 percent of the population between age 25 and 35 has now completed at least a secondary education. Compared to countries included in the recent World Education Indicators, these results are better than most OECD countries. 40 This chapter does not analyze the health system in Albania, because public spending on health as a share of GDP is very small (1.4 percent). 41According to Figure 1, 43 percent of individuals aged 64 or older are illiterate. 35 Figure 4. 1: Improvements in the Education Status of Successive Cohorts. 64 + years 36-64 years M Tertiary 25-35 yer U Secondary 0 Basic 18-24 years O Preschool . Illiterate 14-17 years 6-13 years 0 20 40 60 80 100 Source: Albania Living Standard Measurement Survey, 1996. Hlowever, in Albania it is troubling to note that since 1990 gross enrollment rates have been falling at all levels of education, except tertiary (Table A4.1). For instance, at the upper secondary level, total enrollment rates have declined by almost 50 percent because of huge drops in enrollment in vocational schools. While the large decline in vocational school enrollments may be explained by a shift in rural areas from vocational (agricultural) towards general secondary education, declines in enrollments at other levels of schooling are worrisome. One of the best ways of calculating the impact of declining school enrollments on the population is to consider the "school expectancy" for new entrants. This is the number of years of full-time education that a 6 year-old child can expect to receive during his/her lifetime. Recent calculations suggest that in 1997 school expectancy in Albania was only 42 9.4 years, which is extremely low compared with other countries in the region. For example, school expectancy in other transition economies - such as Hungary and the C'zech Republic - range between 13.9 and 14.4 years. What are the reasons for the falling rates of school enrollment in Albania? On the demand side, an increasingly difficult economic situation has caused large numbers of students at the basic and secondary levels to leave school in order to find work. Indeed, survey data43 suggest that 35 percent of students between the ages of 10 and 14 have left school because of insufficient household income. On the supply side, during the period 1990 to 1992 many primary schools were destroyed because of their close association with the former communist regime. The decreased number of schools - especially in 42 Polomba, Geremia, and Milan Vodopivec. 1999, "Efficiency, Equity and Fiscal Impact of Education in Albania", World Bank, forthcoming. 43 Polomba and Vodopivec, 1999. 36 rural areas --may have reduced enrollment rates by raising the travel costs for those planning to attend school. Declining rates of enrollment have been coupled with reduced public expenditures on education. Between 1990 and 1997 public expenditures on education as a share of GDP fell from 4.2 to 3.3 percent. In real terms, total public expenditures on education in 1997 were less than three-fourths that in 1990. Such trends are worrisome, because public expenditures on education in Albania have traditionally been low compared to other countries in the region. For example, in recent years public expenditures on education as a share of GDP ranged from 6.8 percent in Hungary to 3.9 percent in the Czech Republic. Low and declining public support for education in Albania have adversely affected the quality of schooling through deteriorating material conditions (lack of teaching materials, inadequate building maintenance) as well as non-stimulating conditions for teachers (deteriorating salaries). As the quality of educational services has declined, so has the demand for schooling. A recent survey" found that 18 percent of students between the ages of 10 and 14 have left school because of the deteriorating quality of education. Enrollment Rates and Poverty45 Table 4.1 shows gross enrollment rates by educational level, gender and expenditure quintile group. At the basic level of education (ages 6 to 13), enrollment rates reveal no discernible patterns by gender or income group. In fact, enrollment rates are quite high (above 90 percent) for most income groups. However, at the secondary level (ages 14 to 17) enrollment rates vary on the basis of both gender and income. At the secondary level, both male and female rates of enrollment rise sharply with income group, and female rates of enrollment generally exceed that of males. Even more importantly, at the secondary level gross rates of enrollment are much lower - 60 percent or more -- than those for basic education. Evidently, large numbers of children are dropping out of the educational system after compieting their basic education. This is a troubling development, one which is fueled by both the increasing direct and indirect costs of education as well as tightened economic circumstances. The drop out rates from secondary school are much higher for the poor relative to richer households. Gross enrollment rates drop from 94 percent in basic education to 10 percent in secondary education for the lowest quintile. In the top quintile the dropout rate is 44 Polomba and Vodopivec, 1999. Results from the 1996 LSMS survey in Albania can be used to examine the effects of the educational system on poverty and equity. Since this survey was a single-round. household expenditure study, which collected only very basic data on education,45 it cannot be used to examine the dynamic effects of recent changes in the Albanian educational system. However, the 1996 LSMS can still be used to provide a useful snapshot of the state of that system in 1996. 37 Table 4. 1: Gross Enrollment Rates by Educational Level, Gender and Expenditure Quintile Group. Basic education Secondary education Tertiary education Total per capita Male Female Male Female Male Female -expenditure quintile Lowest 87.7 99.9 11.9 9.0 1.2 2.2 Second 104.9 90.9 24.5 47.2 2.6 2.4 Third 95.3 96.8 27.6 57.8 1.5 8.0 Fourth 93.6 109.8 29.4 25.9 3.4 2.8 Top 101.0 84.5 59.4 60.6 6.6 3.2 Total 95.3 97.4 30.0 38.3 3.3 3.7 Source: Albania Living Standard Measurement Survey, 1996. lower. Gross enrollment rates decline from 94 percent in basic education to 60 percent in secondary school(Table A4.3). A comparison of gross, net and age-specific enrollment rates by level of education and income also shed some light on the prevalence of repeaters the education system. Figure 4.2 shows that the net enrollment for basic education is 82 percent. That is, 82 percent of basic education age students are enrolled in basic education. However, only 25 percent of secondary age students are enrolled in secondary school; Almost 35 percent of secondary school age are enrolled in basic education, or at a level of education that is lower than that dictated by their age. Thus, these students are likely repeaters in the system. Footnote here: Another interpretation is that, these students may be late entrants to the school system.. Grade level information on students (not available in the survey ) provides more accurate analysis of repetition. Figure 4. 2: Net, Gross and Age Specific Enrollment Rates By Level of Education W Basic ESecondary OTertiary 96% 100% - 81% bl/o | . 11 179% Cu 75%- g 50% * 34% ° 25%- 0% _ _I 0% Age Specific Gross Net Level of Schooling Source: Albania Living Standard Measurement Survey, 1996. 38, Table 4. 2: Gross Enrollment Rates by Urban/Rural Sector, Gender and Expenditure Quintile Group Basic education Secondary education Tertiary education Total per capita expenditure quintile Urban Rural Urban Rural Urban Rural Lowest 74.0 96.1 20.3 9.6 18.7 0.9 Second 102.3 94.8 47.6 28.6 6.6 0.9 Third 90.0 99.3 62.7 30.5 7.5 2.7 Fourth 104.9 98.2 42.6 13.2 5.8 1.3 Top 101.1 84.8 75.7 36.4 11.2 0.0 Total 96.8 96.1 57.0 21.2 8.3 1.1 Source: Albania Living Standard Measurement Survey, 1996. At the secondary and tertiary levels, why are enrollment rates for the poor so much lower than those of the rich? Studies in other countries generally attribute low enrollment rates of the poor to high out-of-pocket fees (clothes, books, etc), distance to school, and the opportunity costs of foregone wages. In the case of Albania, these and other factors mean that at the secondary level the net enrollment rate of the poor is less than 10 percent. Enrollment rates also vary between urban and rural areas (Table 4.2). While gross rates of enrollment are fairly similar at the basic level of education, they differ widely at the secondary and tertiary levels. At the secondary level, only 21 percent of all rural children of secondary school age are in school as compared to 57 percent of those in urban areas. The enrollment gap is even wider at the tertiary level. Only 1 percent of all rural children of tertiary school age are in school as compared to 8 percent of those in urban areas. Lower gross enrollment rates in rural areas reflect two factors. First, most secondary and tertiary schools are located in urban - and not rural -- areas. This makes it more expensive for rural households to send their children to school. Second, the opportunity costs of schooling (in the form of foregone wages) as a percent of total household income are higher in rural as opposed to urban areas. The combination of these two factors makes it particularly difficult for poor rural households to keep their children to school. In Table 4.2 the gross enrollment rate for the rural poor at the secondary school level - 9.6 percent - is among the lowest in the table. Education and Equity In 1996 the Government spent nearly 3.4 percent of GDP on education. The public education system is almost completely funded by the Government. Therefore, spending per student can be calculated from the 1996 LSMS by taking total recurrent expenditures 39 Table 4.3: Per Capita and Total Expenditure (lek) on Basic, Secondary and Tertiary Education, Albania 1996. Basic Secondary Tertiary Total per capita Per capita Percent of Per capita Percent of Per capita Percent of expenditure expenditure total expenditure total expenditure total quintile (lek) expenditure (lek) expenditure (lek) expenditure Lowest 5,230 27.0 2,423 7.2 2,394 7.5 Second 5,429 18.8 7,855 21.2 3,503 16.0 Third 5,345 21.8 10,342 24.7 6,324 26.7 Fourth 5,591 20.5 6,443 14.8 4,445 18.2 Top 5,231 11.8 13,944 32.2 7,157 31.6 Total 5,425 100.0 7,907 100.0 4,892 100.0 Source: Albania Living Standard Measurement Survey, 1996. on education at each level of education (less educational expenses in Tirana)46 divided by the gross number of students enrolled at each level of education. This methodology, which assumes that public spending is the same for all students at a given level of education (e.g. basic), has certain problems,47 but is the best that can be done in the absence of dis-aggregated data on the pattern of public expenditures on education. In 1996 annual spending per student in Albania averaged (in nominal terms): 5,568 lek at the basic level, 23,272 lek at the secondary level and 140,918 lek at the tertiary level. Initially, these figures suggest that Government expenditures are biased towards the upper levels of education, which are usually not very accessible to the poor. This is a pattern frequently observed in other developing countries. However, it should be noted that since far more students are enrolled in basic - as opposed to tertiary - education,48 Govemment expenditures on education are more pro-poor than they may seem. In 1996 the Government actually spent a larger share of its educational budget on basic education (38.3 percent) as opposed to either secondary (26.3 percent) or tertiary education (24.5 percent).49 Table A4.2 shows that at the basic level of education, Government spending is biased towards the poor. Poor students (those in the lowest quintile group) of both sexes - male and female -- receive a larger share of total expenditures on education than do those in the top quintile group. The main reason for this is that far more poor children are enrolled in 46 Tirana is excluded from the calculations because the 1996 LSMS did not include households in the capital city of Tirana. 47 For example, in most countries public spending on education is higher in urban as opposed to rural areas. However, without district-specific data on the distribution of public spending between urban and rural schools, it is impossible to analyze the extent of the urban bias in public spending on education in Albania. 48 In 1996 525,900 students were enrolled in basic education, 116,100 in secondary education and 32,600 in tertiary education. 49 In 1996 the Government also spent 10.9 percent of its educational budget on pre-school (ages 3 to 5). 40 basic education than children from the upper income groups. However, even the pattern of expenditures in Table 4.2 seems to be progressive because per capita expenditures on education do not seem to rise much with income group. These results, however, do not hold at the secondary level. Table A4.3 shows that at the secondary level poor students of both sexes receive a much smaller share of total public expenditures on education. At the secondary level boys in the lowest quintile group receive about 1/5 of the amount of per capita expenditures as their counterparts in the top quintile group, and poor females receive even less. The level of per capita expenditures on education at the secondary level also rises sharply with income group. These differences in the incidence of benefits between basic and secondary school reflect the impact of changes in the proportion of children enrolled at different levels. At the lower levels of education, gross enrollment rates are high for all quintile groups, including the poor. But as children mature, and begin to drop out of school, enrollment rates decline most dramatically for the poor. For example, gross enrollment rates for boys in the lowest expenditure quintile group fall from 88 to 12 percent between basic and secondary education, while enrollment rates for boys in the top quintile group fall from 101 to 59 percent (Table 4.1). Table A4.4 completes the analysis by showing the pattern of public expenditures at the tertiary level of education. As at the secondary level, students in the top income groups receive the bulk of the share of public expenditures on education. At the tertiary level males in the top expenditure quintile group receive almost 9 times the share of expenditures on education as those in the lowest quintile group. Increased public investment in the education of the poor would do much to help alleviate long-term poverty in Albania. Multivariate analysis shows that returns to education are high in Albania, though only in the public sector.50 However, as in other transition economies, over time, there are likely to be higher returns to education in the private sector. Particularly if policies to promote the growth of non-agricultural private sector are followed. For these reasons, investment in education (especially in rural areas) should help reduce poverty by increasing the ability of graduates to find and retain skilled work. In Albania the exact type of investment in education (e.g., infrastructure, loans or scholarships for the poor to cover out of pocket costs) should be guided by an effort to reduce the large dropout rates of the poor. From the standpoint of long-term poverty alleviation, some means needs to be found to keep more poor children of both sexes in secondary and tertiary education. This would focus public spending more heavily on the poor. 500n Slovenia, see Peter Orazem and Milan Vodopivec, "Winners and Losers in Transition: Returns to Education, Experience and Gender, World Bank Economic Review (May 1995); on Estonia, see Rivo Noorkoiv et al., "Employment and Wage Dynamics in Estonia," Economics of Transition (1998). 41 B. Public Transfers This section discusses the main public transfer programs in Albania and their potential for alleviating poverty. With the virtual elimination of food and other subsidies in Albania,5' only six types of public transfer programs remain: (1) old age pensions; (2) Ndihme social assistance; (3) unemployment benefits; (4) social welfare payments; (5) disability and caretaker transfers; and (6) other pensions. In 1996, the Government spent almost 8.5 percent of GDP on public transfer programs. The largest share of government outlay on transfers went to pensions, which accounted for 6 percent of GDP. Spending on Ndihme social assistance accounted for 1.4 percent of GDP, and spending on all other transfer programs accounted for 1.1 percent of GDP. The main purpose of Government spending on public transfers is to reduce poverty and improve equity. This section measures the ability of these transfer programs to meet these goals by focusing on their efficiency and effectiveness. The efficiency of transfer programs is evaluated on the basis of their ability to deliver benefits to the target population (the poor) and to exclude the non-target population (the non-poor). In other words, the efficiency of transfer programs is measured on the basis of their ability to avoid the leakage of benefits to the non-poor. The effectiveness of transfer programs is judged by the size of the transfer received by the poor. Transfer programs are judged to be pro-poor (progressive) if the share of benefits received by the poor is larger than their proportion of the population. As in the previous sections, all of these evaluations are based on expenditure as a proxy for income. An Overview of Public Transfers Figure 4.3 shows the share of total transfers going to each program. In 1996, almost 70 percent of total public transfers in Albania went to old age pensions. The next largest share (12 percent) of transfers goes to the Ndihme social assistance program. The Ndihme program is a means-tested program designed to provide income support to families with low household earnings or inadequate landholdings. Each of the other transfer programs - unemployment, social welfare, disability and caretaker - account for comparatively small shares (each program less than 10 percent) of total public transfers. T'able A4.6 shows the distribution of public transfers by expenditure quintile group. In 1996, the Ndihme social assistance is the most progressive (pro-poor) because it delivers the highest level of benefits to the poor (those in the bottom quintile group). Not only do transfers from the Ndihme program account for almost 5 percent of the total expenditures of those in the lowest quintile group, but transfers from this program as a percent of total household expenditures fall as income rises. Subsidies have been cut from over 20 percent of GDP in 1991 to only about 0.5 percent in 1997. The only remaining subsidies in Albania are for school books, passenger and railway transport, drinking water, water for irrigation and funeral expenses. 42 By contrast, transfers from the most important program in Albania -- old age pensions -- actually increase with income (Table A4.6). Transfers from old age pensions rise from 3.7 percent of total expenditures for the poor to 7.4 percent of such expenditures for the richest group. Old age pensions thus appear to have a regressive impact on poverty and inequality. While unemployment benefits represent a small share (1.1 percent) of total expenditures of the poor, Table A4.6 shows that these benefits fall as total expenditures rise. Unemployment benefits thus seem to have a progressive impact on equity. These benefits were, in fact, designed to benefit those people who had just lost their job. The Albanian system of unemployment insurance and benefits was instituted in July 1992 in order to cope with the rising tide of unemployment in that country. Registered unemployment peaked in Albania in 1992 at 27 percent and has since fallen to about 13 percent. Very few of the registered unemployed now receive benefit. Figure 4. 3: Public Transfers in Albania as Percent of Total Public Transfers. 4%4 Oa57 M Old Age Pension * Other Pensions iS 19IIII %II o Unemployment Benefits o Ndlrhme Social Assistance * Social Welfare Paymnents 68 49% D oisability AllowancelCaretakerGrantlin Kind Transfers Source: Albania Living Standard Measurement Survey, 1996. Transfers from all other programs in Albania are quite small, accounting for less than 2 percent of total expenditures for each quintile group (Table A4.6). For this reason, the following discussion will focus on the two leading transfer programs in Albania: the Ndihme social assistance program and old age pensions. 43 Ndihme Social Assistance Program As in much of Eastern Europe, the breakdown in central planning in Albania in the early 1990s caused many industrial and business enterprises to close. This led to both massive layoffs and steadily rising unemployment. In an effort to assist laid-off workers, in July 1993 the Government created a new safety net program, the Ndihme Ekonomika (NE) program, which was designed to provide income support to households falling below a 52 certain minimum threshold. The Ndihme program was designed to assist urban families with no other source of income and to support rural families with small and inadequate landholdings. Initially, the Ndihme social assistance program was an entitlement to families conditional on an income criterion that was set in the Albanian capital, Tirana. The Government established minimum and maximum levels for the grants and monitored their distribution to local units. Families applied for Ndihme benefits at the local offices of the Ministry of Labor and Social Protection (MOLSP), and had to provide documentation of their landholdings and current employment status. Table 4. 4: Ndihme Social Assistance: Households Receiving Benefits Per Capita Urban Rural Expenditure Percent of HHs in decile Percent of HHs in decile Decile receiving Ndihme social receiving Ndihme social assistance assistance 1 56.5 41.2 2 19.6 27.7 3 18.9 31.4 4 19.7 19.8 5 16.0 14.2 6 11.3 14.8 7 20.0 6.8 8 11.6 7.3 9 1.0 5.5 10 2.5 6.5 Total 11.2 17.8 Source: Albania Living Standard Measurement Survey, 1996. In 1995 the Ndihme program was reformed in order to increase the efficiency of the targeting of benefits. In that year local administrators (municipalities in towns and communes in rural areas) were given more powers to review the distribution of Ndihme benefits. Specifically, local administrators were allowed to retain 50 percent of the difference between the amount of social assistance distributed by the MOLSP and the amount allocated to the community. The money that local administrators was allowed to keep could be used for community projects. 52 For more details on the Nhihme social assistance program, see Harold Alderman, "Decentralization and Targeted Transfers: Social Assistance in Albania," (forthcoming). 44 In the case of the Ndihme program, while its targeting capacity may well have been reduced as a result of the recent crisis, in 1996, the combined use of means-testing and local community review seems to have produced a moderately targeted transfer program. Table 4.8 shows the percent of households receiving Ndihme social assistance by expenditure decile group. If there were perfect targeting, and only the poor (those in the bottom two deciles) were receiving benefits, then the entries for the two lowest groups would be 100 percent. However, the results show that targeting is far from perfect: many poor households are excluded from the Ndihme program because the entries for the bottom two decile groups range between 27 and 56 percent. On the other hand, the Ndihme program appears to be at least moderately targeted to the poor. For example, when the bottom two deciles are combined together, 32 percent of poor urban households and 35 percent of poor rural households are receiving benefits from the Ndihme program (Table 4.8). Moreover, the percentage of households receiving benefits falls as income increases. In the top decile group less than 7 percent of urban or rural households receive Ndihme program benefits. The Ndihme program also seems to be moderately targeted to the poor when the effectiveness of the program is considered. Table 4.8 shows the share of total Ndihme program benefits going to different expenditure decile groups. In urban areas the Ndihme program does not seem to be effectively targeted to the poor because those in the bottom two deciles do not receive their proportionate share (10 percent) of program benefits.53 However, in rural areas the situation is reversed and the Ndihme program does seem to be pro-poor. In rural areas each of the bottom two deciles receive more than their proportionate share (10 percent) of Ndihme program benefits. Since it is moderately targeted to the poor, the Ndihme program also has a positive impact on inequality. This can be seen by comparing changes in the Gini coefficient of inequality54 between two situations: when benefits from the Ndihme program are excluded from total per capita expenditures, and when such benefits are included in expenditures. In urban areas, when benefits from the Ndihme program are included in expenditures, the Gini coefficient falls by 2.8 percent: from 0.257 to 0.250. In rural areas, when Ndihme program benefits are included the Gini coefficient falls by 2.5 percent: from 0.283 to 0.276. In other words, in both urban and rural Albania benefits from the Ndihme program help to improve the overall distribution of income. 53 These results are based on the 1996 LSMS which, as noted above, did not include the capital city of Tirana. 54 The Gini coefficient of inequality is a commonly used measure of income inequality. It is scaled to lie between zero (perfect equality) and 1 (perfect inequality). 45 Old Age Pensions Old age pensions represented the single largest category (43 percent) of social spending in Albania in 1996, and accounted for 17 percent of total government spending and around 6 percent of GDP. Pensions have at least two important effects on the economy. First, the pension system has been running persistent deficits which have contributed to the high fiscal deficits in Albania at large.55 Second, while pensions are in fact deferred savings that typically go to higher income households,.56 pension benefits are often cited as an important source of income from the old and disabled, who typically find it more difficult to adjust to the movement from socialism to a market economy. Table 4.5 shows the percent of households receiving old age pensions by expenditure decile group. Given their savings objectives, the results show that pensions are not targeted to the poor. In fact, most poor households do not even receive pensions: the percentage of households receiving pensions in the bottom two decile groups (urban and rural) never exceeds 30 percent. Moreover, the percentage of households receiving Table 4. 5: Old Age Pensions: Households Receiving Beneflts Per Capita Urban Rural Expenditure Percent of households in Percent of households in Decile decile receiving old age decile receiving old age pensions pensions 1 10.5 22.5 2 10.3 29.6 3 24.3 31.1 4 35.1 33.3 5 37.7 47.7 6 30.9 39.5 7 26.0 25.4 8 35.2 32.8 9 36.4 38.7 10 45.8 43.8 Total 34.5 34.1 Source: Albania Living Standard Measurement Survey, 1996. pensions rises steadily with income group. In fact, between two and four times as many households in the top decile group receive pensions as do those in the bottom decile group. Since pensions are mainly deferred savings, and upper-income groups tend to save more than the poor, pensions in Albania tend to accrue to the rich. 55 In 1997 the fiscal deficit/GDP in Albania was -11.9. 56In 1997 more than 15 percent of the population received pensions. 46 Old age pensions also seem to go mainly to the rich in Albania when the effectiveness of the program is considered. Table A4.8 which shows the share of pensions going to different decile groups, reveals that in urban areas the bottom two deciles receive less than 1 percent of total pension benefits. In rural areas the poorest two deciles receive less than 10 percent of total pension benefits. In both urban and rural areas the largest share of pension benefits goes to those in the highest decile group. Because of the way that they are targeted, old age pensions tend to have a slightly negative impact on inequality. For example, in urban areas when benefits from pensions are included in total per capita expenditures, the Gini coefficient of inequality rises by a slight 0.3 percent: from 0.257 to 0.258. In rural areas, however, when pension benefits are included in expenditures, the Gini coefficient falls by 1.1 percent: from 0.283 to 0.280. This difference is probably due to the fact that in rural areas the poorest deciles receive a higher proportion of pension benefits than do their counterparts in urban areas (Table A4.8). In summary, public spending on social protection-mainly benefits richer households. This is mainly because public pensions, which consume the largest share of public social protection spending, because of their link with an individuals past wages, are received mainly by the better off; Moreover, the largest share of government spending on education is pro-rich, as children enrolled secondary and higher education belong to higher income households. The only pro-poor public programs are compulsory basic education, in which the poor participate disproportionately more than the rich; unemployment insurance and, the main poverty alleviation program, social assistance. However, public spending on these programs is much smaller relative to pensions and secondary and higher education. 47 V. REFERENCES AND BIBLIOGRAPHY Alderman, Harold. 1999. Decentralization and Targeted Transfers: Social Assistance in Albania. World Bank. Washington DC (forthcoming). La Cava, et. al. 1999. Albania: Filling the Vulnerability Gap, World Bank, Washington DC. Processed. Orazem, Peter and Milan Vodopivec, May 1995. Winners and Losers in Transition: Returns to Education, Experience and Gender, World Bank Economic Review. Noorkoiv, Rivo et. al. 1998. Employment and Wage Dynamics in Estonia. Economics of Transition. Polomba, Geremia, and Milan Vodopivec. 1999. Efficiency, Equity and Fiscal Impact of Education in Albania. World Bank, forthcoming. World Bank. 1996. Albania: Growing out of Poverty. Report No. 15698. Washington DC. World Bank. 1996. Growing Out of Poverty. The World Bank, Washington DC. World Bank. 1997. Country Economic Memorandum. Washington DC. Processed. World Bank. 1997. Romania, Poverty and Social Polity. Report No. 16462-RO. Washington DC. Tang, et. al. 1998. Country Economic Memorandum: Albania. World Bank, Washington DC. World Bank. 1998. Kazakhstan, Living Standards During Transition. Report No.7520- KZ. Washington DC. World Bank. 1998. World Development Indicators. Washington DC. World Bank 1999. Bulgaria, Poverty During the Transition. Report No. 18411. Washington DC. World Bank. 1999. Former Yugoslav Republic of Macedonia: Focusing on the Poor. Report No. 18411. Washington DC. World Bank 1999. Kyrgyz Republic, Update on Poverty in the Kyrgyz Republic. Report No. 19425-KG. Washington DC. 48 STATISTICAL ANNEX 49 Table of Contents for Annex Tables and Figures Tables Table A2. 1: Mean Characteristics of the Sample by Location ....................................................... 53 Table A2. 2: Mean Characteristics of the Sample Region and Area ............................................... 54 Table A2. 3: Mean Characteristics of the Sample by Quintile ...................................................... 55 Table A2. 4: Actual Numbers of Migrants the Analysis is Based On (Number of Individuals)....56 Table A2. 5: Average Remittances Received by Each Type of Household ................................... 56 Table A2. 6: Poverty and Inequality Indicators, Albania 1996' ................................................... 57 Table A2. 7: Probit Regressions of Characteristics of those Above and Below the Poverty Line, Albania 1996 ( Probability at the Individual Level) ................................................. 59 Table A2. 8: Average Area of Land Owned by Type Given in Dynyms (Number of Households Based On) in Albania 1996 .................................................. 61 Table A2. 9: Average Area of Land Owned by Type Given in Dynyms (Number of Households Based On) in Albania 1996 ...................................................... 62 Table A2. 10: Average Land Area and Value in Rural Areas ..................................................... 62 Table A2. 11: Average Value of Land Owned by Type for Households that Owns Each Type of Land in Albania 1996 .................................................. 63 Table A3. 1: The Composition of the Population, Active Labor Force, Employment and Unemployment, Albania ................................................. 66 Table A3. 2: Socio-Demographic Characteristics of Labor Force Participation Rate, Unemployment Rate and Long Term Unemployment Rate ................................................. 67 Table A3. 3: Socio-Demographic Composition of the Unemployed: Never Worked (NWU), Short-term (STU), and Long-term (LTU) Unemployed ................................................. 68 Table A3. 4: Socio-Demographic Composition Public and Private Sector Employment .............. 69 Table A3. 5: The Profile of New Hires (as Defined by Elections), Albania ................................... 70 Table A3. 6: The Profile of New Hires (as Defined by Tenure), Albania . ..................................... 71 Table A3. 7: Estimates of Human Capital Earnings Functions (OLS) - Albania ........................... 72 Table A3. 8: Average and Median Net Monthly Wage, Albania ................................................... 72 Table A3. 9: Sector of Employment by Industry, Albania ............................................................. 74 Table A3. 10: Sector of Employment by Industry, Albania ........................................................... 74 Table A3. 11: Type of Employment Contract by Industry, Albania .............................................. 75 Table A3. 12: Type of Employment Contract by Industry, Albania .............................................. 75 Table A3. 13: Sector and Length of Employment Contract by Industry, Albania ........................ 76 Table A3. 14: Summary of Earnings Distribution by Sector, Albania ........................................... 76 Table A3. 15: The Incidence and Composition of Low-Paid Employment, Albania .................... 77 Table A3. 16: Contribution of Selected Variables to Log-Earnings Inequality ............................. 78 Table A4. 1: Gross Enrollment Rates by Level of Education, 1990 to 1997 ................................. 79 Table A4. 2: Public Expenditures on Basic Education by Gender and Expenditure Quintile Group ............................................................................................................................................... .79 Table A4. 3: Public Expenditures on Secondary Education by Gender and Expenditure Quintile Group ...................................................................... 80 Table A4. 4: Public Expenditures on Tertiary Education by Gender and Expenditure Quintile Group ...................................................................... 80 Table A4. 5: Net, Gross and Age-Specific Enrollment Rates by Educational Level and Expenditure Quintile Group ....................................................................... 81 Table A4. 6: Distribution of Public Transfers by Expenditure Quintile Group ............................. 81 Table A4. 7: Age specific, Gross and Net Enrollment Composition of 6 to 24 Year Olds, Albania 1996 .................................................................. 82 51 Table A4. 8: Old Age Pensions: Households Receiving Benefits . ................................................ 83 Table A4. 9: Percent of Total Per Capita Assistance from Old Age Pensions Received by Decile ............................................................................................................................................... .83 Figures Figure A2. 1: Distribution of income by source, Albania 1996 ..................................................... 60 Figure A2. 2: Distribution of Income by Source of Income . ......................................................... 61 Figure A2. 3: Per Capita Consumption Density by Region, Albania 1996 .................................... 63 Figure A2. 4: Per Capita Consumption Distribution by the Age of the Household Head . ............ 64 Figure A2. 5: Per Capita Consumption Density by the Employment Status of Household Head. 64 Figure A2. 6: Cumulative Distribution of Consumption in Urban and Rural Areas,. Albania 1996. ..... ......................................................................................................................................... 65 Figure A2. 7: Cumulative Distribution of Consumption in the North and South, Albania 1996... 65 52 Table A2. 1: Mean Characteristics of the Sample by Location. Sample North/ North/ South/ South/ (7603) Urban Rural Urban Rural (696) (3450) (1066) (2391) Household size (present 5.38 4.99 6.00 4.37 5.73 members only) Household size (with absent 5.61 5.11 6.20 4.55 6.03 members, n=7603) Age of household head Mean age 49.3 47.4 49.5 49.2 50.7 [Minimum, Maximum] [17,96] [24,85] [17,96] [19,86] [20,95] Mean number of members per household in each age group 0-15 years (Number of 2.11 1.76 2.62 1.49 2.25 children) 16-35 years 1.84 1.65 2.01 1.39 2.06 36-64 years 1.36 1.50 1.32 1.37 1.34 64+ years 0.31 0.20 0.25 0.31 0.39 Education of the Household Head Less than Basic 10.97 0.45 10.45 7.77 15.85 Basic 55.06 41.85 64.08 40.23 61.74 Secondary 28.62 40.79 22.54 41.43 21.51 Tertiary 5.35 16.91 2.92 10.57 0.90 Presence of any migrants No mnigration 90.13 86.54 88.93 91.16 91.10 Returned home 1.39 2.71 2.18 0 1.47 In Albania 2.09 1.70 1.60 3.20 1.76 Abroad 6.76 9.47 8.29 5.64 5.85 Migration status (head) No migration 98.51 97.08 97.88 100 98.33 Returned home 0.71 1.64 0.84 0 0.88 In Albania 0.13 0.00 0.12 0 0.26 Abroad 0.65 1.28 1.23 0 0.54 Employment (Mean number of members) Wage 0.43 0.70 0.28 0.68 0.32 Self employed 0.15 0.30 0.10 0.24 0.08 Agriculture 1.30 0 1.67 0.10 2.16 Unemployed 0.54 0.95 0.35 1.04 0.24 Retired 0.33 0.45 0.24 0.48 0.25 Other 0.76 0.96 0.94 0.53 0.74 Per Capita Consumption Mean (Variance) 4960 5523 3678 6125 4936 (2666) (3040) (1979) (2624) (2621) Skewness (Kurtosis) 2.07 4.09 1.80 (7.4) 0.90 (3.5) 2.57 (13.71) (36.3) (16.2) In calculating per capita consumption only the individuals that were present were used for the household size. Then this per capita consumption level was assumed to be that of the absent members as well. In other words absent members were not dropped from the sample. 1 This does not necessarily add to 100 given certain households have 2 types of migTants (I individual that has returned home as well as one that is currently living abroad). 53 Table A2. 2: Mean Characteristics of the Sample Region and Area. Rural Urban North South Household size (present members 5.84 4.52 5.74 5.18 only) Household size (with absent 6.10 4.71 5.92 5.44 members, n=7603) Age of household head NMean age 49.5 48.9 47.8 50.1 [Minimum, Maximum] [19,86] [17,96] [17,96] [19,95] Mean number of members per household in each age group 0-15 years (Number of children) 2.40 1.56 2.40 1.94 16-35 years 2.04 1.46 1.92 1.79 36-64 years 1.33 1.40 1.37 1.35 64+ years 0.33 0.28 0.24 0.36 Education of the Household Head Less than basic 13.66 5.89 7.94 12.63 Basic 62.69 40.64 58.50 53.18 Secondary 21.93 41.27 27.12 29.44 ITertiary 1.72 12.20 6.43 4.75 Education of the Individual Less than basic 8.42 4.42 6.70 7.07 Basic 65.36 38.66 56.07 55.27 Secondary 24.30 47.05 31.60 33.22 Tertiary 1.92 9.87 5.63 4.44 Migration status (head) No migration 98.11 99.25 97.62 98.99 Returned home 0.87 0.42 1.04 0.53 In Albania 0.20 0.00 0.09 0.16 Abroad 0.82 0.33 1.24 0.32 Employment (Mean number of members) Wage 0.30 0.68 0.38 0.46 Non wage non agriculture 0.09 0.26 0.15 0.15 Non wage agriculture 1.96 0.07 1.25 1.33 IJnemployed 0.29 1.01 0.50 0.56 Retired 0.25 0.48 0.30 0.34 Other 0.82 0.64 0.94 0.65 Per Capita Consumption Mean (Variance) 4426 (2460) 5970 4141 (2427) 5409 (2685) (2749) Skewness (Kurtosis) 2.40 (15.0) 1.96 (15.3) 3.2 (29.8) 1.77 (9.7) In calculating per capita consumption only the individuals that were present were used for the household size. Then this per capita consumption level was assumed to be that of the absent members as well. In other words absent members were not dropped from the sample. 54 Table A2. 3: Mean Characteristics of the Sample by Quintile Poorest 2nd Quintile 3rd Quintile 4th Quintile Richest Household size (present 6.9 6.2 5.3 4.6 4.0 members only) Household size (with absent 7.0 6.3 5.5 4.9 4.3 members, n=7603) Age of household head Mean age 47.4 49.5 51.3 48.3 50.1 [Minimum, Maximum] Mean number of members per household in each age group 0-15 years (Number of 3.4 2.4 1.9 1.7 1.2 children) 16-35 years 1.9 2.2 1.8 1.7 1.6 36-64 years 1.4 1.4 1.4 1.3 1.3 64+ years 0.3 0.3 0.4 0.2 0.3 Education of the Household Head Illiterate 11.0 7.5 8.5 6.2 3.8 Pre-School 2.0 5.8 4.7 2.5 1.6 Basic 68.6 55.1 57.1 50.2 42.7 Secondary 17.9 26.4 26.2 34.8 40.1 Tertiary 0.5 5.2 3.6 6.3 11.9 Employment (Mean number of members) Employed (wage/non wage) 2.1 2.0 1.7 1.8 1.7 Unemployed 0.4 0.7 0.7 0.5 0.4 Retired 0.2 0.3 0.4 0.3 0.4 Other 0.8 1.0 0.7 0.6 0.7 Education of Individual Less than basic 8.88 7.54 8.84 4.44 5.69 Basic 72.83 59.95 55.61 52.97 41.97 Secondary 17.36 29.75 31.83 37.08 42.61 Tertiary 0.93 2.78 3.72 5.51 9.73 Per Capita Consumption Mean (Variance) 2238 3358 4361 5755 8911 (436) (278) (302) (481) (2477) Skewness (Kurtosis) -0.70 0.04 (1.9) 0.32 (2.1) 0.23 3.57 (27.3) (3.0) (1.86) 55 Table A2. 4: Actual Numbers of Migrants the Analysis is Based On (Number of Individuals). Returned Home In Albania Abroad Any migrant in household North 20 (118) 22 (90) 57 (354) South 12 (68) 22 (105) 33 (184) Urban 27 (167) 32 (151) 62 (416) Rural 5 (19) 12 (44) 28 (122) Total 32 (186) 44 (195) 90 (538) Migrant household head North 9 (51) 1 (7) 14 (75) South 8 (37) 2 (14) 3 (17) Urban 14 (77) 3 (21) 15 (83) Rural 3(11) 0(0) 2(9) Total 17 (88) 3 (21) 17 (92) Note: The numbers in this table are not weighted, but given to get a sense of the observations the analysis of migrants is based on. In parenthesis is the total number of individuals from households that report each type of migrant sending remittances. Table A2. 5: Average Remittances Received by Each Type of Household. Returned Home In Albania Abroad Any migrant in household North 23,985 (55.7) 8,106 (51.5) 10,720 (49.9) South 10,305 (63.3) 2,719 (44.9) 10,580 (54.4) Rural 20,074 (61.2) 6,161 (49.9) 11,286 (54.3) Urban 12,166 (44.1) 3,416 (43.7) 9,303 (46.0) Total 18,838 (58.6) 5,412 (48.2) 10,669(51.5) Household head North 34,444 (60.3) 0 (0) 8,291 (55.2) South 6,083 (51.0) 7,666 (74.5) 0 (0) Rural 21,452 (55.2) 5,111 (49.7) 4,261 (44.1) Urban 19,444 (59.6) - (-) 26,083 (100) Total 21,098 (55.9) 5,111 (49.7) 6,828 (51.5) Note: The numbers in this table are not weighted, but given to get a sense of the observations the analysis of migrants is based on. In parenthesis is the total number of individuals from households that report each type of migrant sending remittances. 56 Table A2. 6: Poverty and Inequality Indicators, Albania 19961 At the Household Level Number As % of As % of HC Index Poverty Poverty Poverty Gini of Poor the Poor Total (% poor) Gap Index Gap (%) Severity Index ('000 S)2 Population (%) Index Total 517.1 100 100 20.00 4.49 22.44 1.46 0.276 Rural 465.7 90.1 65.4 27.54 6.38 23.15 2.10 0.277 Urban 51.5 9.9 34.6 5.76 0.92 15.98 0.26 0.242 North 301.3 58.3 35.4 32.90 8.14 24.74 2.86 0.286 South 215.9 41.7 64.6 12.93 2.49 19.23 0.70 0.258 North/Urban 13.9 2.7 8.9 6.07 1.07 17.63 0.34 0.253 North/Rural 287.3 55.6 26.5 41.89 10.51 25.09 3.70 0.273 SouthUrban 37.5 7.3 25.7 5.65 0.87 15.36 0.23 0.236 SouthlRural 178.3 34.5 38.9 17.75 3.56 20.04 1.01 0.261 Male headed households 470.3 91.0 90.6 20.1 4.34 21.63 1.37 0.271 Female headed households 46.8 9.0 9.4 19.3 5.89 30.58 2.36 0.313 Education (Household Head) Lessthanbasic 68.3 13.2 11.0 24.10 6.32 26.24 2.22 0.316 Basic 353.3 68.3 55.1 24.82 5.49 22.13 1.78 0.272 Secondary 92.7 17.9 28.6 12.53 2.61 20.86 0.81 0.254 Tertiary 2.8 0.5 5.4 2.03 0.42 20.88 0.09 0.228 Household size 1-4 members 56.2 10.9 30.8 5.89 1.06 18.04 0.31 0.248 5-7 members 297.8 57.6 49.2 23.27 5.22 22.43 1.62 0.243 8-10 members 125.2 24.2 16.0 45.98 12.15 26.42 4.64 0.261 11-16 members 37.9 7.3 4.0 48.01 7.65 15.93 1.84 0.107 Number of Children in Household 0-1 children 58.0 11.2 41.3 7.1 1.2 17.2 0.3 0.261 2-4 children 317.5 61.4 55.3 20.8 4.2 20.2 1.2 0.248 5-10 children 141.6 27.4 3.4 58.9 17.4 29.5 6.9 0.217 Number of aged in household 0 aged 395.1 76.4 70.0 20.6 4.6 22.6 1.5 0.271 1 aged 95.1 18.4 21.9 18.8 4.2 22.5 1.3 0.276 2 aged 26.8 5.2 8.2 16.7 3.4 20.2 1.1 0.328 Pension receiver In household 148.2 28.7 23.5 15.7 3.2 20.2 1.0 0.281 Not in household 368.9 71.3 76.5 22.5 5.3 23.4 1.8 0.272 Migration Status (Head) No migration 507.8 98.2 97.4 19.42 4.44 22.29 1.34 0.276 Retumed home3 0 0 1.2 0 0 0 0 0.231 In Albania 1.8 0.4 0.3 52.83 4.79 9.07 0.43 0.040 Abroad 7.5 1.5 1.2 44.77 16.18 36.13 7.42 0.368 Migration Status (Presence of any) No migration 479.3 92.7 88.7 20.57 4.66 22.67 1.53 0.269 Returned home 2.6 0.5 2.5 7.17 0.38 5.34 0.02 0.241 In Albania 9.2 1.8 2.6 17.05 2.21 12.94 0.34 0.288 Abroad 27.9 5.4 7.1 15.94 3.56 22.25 1.19 0.302 Age of head 16-35 years 102.3 19.8 17.8 22.69 5.13 22.60 1.77 0.274 36-64 years 350.2 67.8 66.1 21.03 4.71 22.39 1.50 0.269 64+ years 64.6 12.5 16.1 13.78 3.09 22.44 1.03 0.294 Occupation (Household 57 At the Household Level Number As % of As % of HC Index Poverty Poverty Poverty Gini of Poor the Poor Total (% poor) Gap Index Gap (%) Severity Index ('000 s)2 Population (%) Index Head) Wage employed 45.6 8.8 18.4 9.57 1.90 19.87 0.56 0.235 Self employed 7.8 1.5 7.4 4.10 0.62 15.21 0.15 0.235 Agriculture 300.7 58.2 36.6 31.77 7.20 22.67 2.26 0.279 Other not in labor force 47.2 9.4 7.1 25.88 6.58 25.41 2.43 0.290 Unemployed 37.7 7.3 10.2 14.35 3.02 21.04 0.99 0.234 Absent 16.6 3.2 3.0 21.37 5.29 24.76 2.36 0.289 Retired 61.4 11.9 17.3 13.73 3.03 22.09 1.04 0.286 'The lower poverty line is defined to be the highest per capita consumption (2,888 lek) for the 1st quintile of households. 2Actual number in the category after inflating with the sample weights. 3 Incidence of poverty is 0 among households where the head has retumed. 58 Table A2. 7: Probit Regressions of Characteristics of those Above and Below the Poverty Line, Albania 1996 (Probability at the Individual Level). Model 1 Model 2 Coefficient (SE) Coefficient (SE) Albania Rural Areas Individual Characteristics Age of individual ' 0 to 5 years -0.02 -0.05 6 to l5years -0.02 -0.04 16 to 35 years -0.001 -0.01 In school -0.03 * -0.04 Education of the individual' Basic 0.02 0.02 Secondary -0.02 -0.05 Tertiary -0.02 -0.01 Employment status of the individual 2 Wage employment -0.06 *** -0.10 *** Agriculture -0.03 ** -0.06 Self employed -0.08 ** -0.14 ** Unemployment -0.02 -0.04 Absent -0.07 *** -0.13 *** Pension -0.04 ** -0.05 Characteristics of the household Household size 0.01 *** 0.02 *** Number of children Aged 0-5 years 0.07 *** 0.09 *** Aged 6-15 years 0.07 *** 0.10 *** Characteristics of the household head Female head -0.01 -0.08 *** Age of head Age 16 to 35 years 0.23 *** 0.35 *** Age 36 to 64 years 0.09 *** 0.20 *** Education of the household head 3 Basic -0.10 *** -0.23 *** Secondary -0.12 *** -0.21 *** Tertiary -0.12 *** -0.21 *** Employment status of the household head 4 Wage employment -0.09 *** -0.18 *** Agriculture -0.03 ** -0.09 *** Self employed -0.12 *** -0.20 *** Unemployment -0.04 ** -0.10 *** Absent -0.05 ** 0.07 * Retired -0.04 * -0.08 ** Presence of a migrant in the household Migrant has returned -0.10 *** -0.19 *** Migrant in Albania -0.02 -0.01 Migrant in foreign country -0.03 * -0.08 *** Location of the household 5 South urban 0.01 -0.22 * North rural 0.20 *** South rural 0.02 * Sample size 7603 5841 R squared 0.3221 0.2787 59 Model 1 Model 2 Coefficient (SE) Coefficient (SE) Albania Rural Areas Reference category is individuals aged 36 years and above. 2Reference category is household heads household heads aged 64 years and over. 3 Reference category is household heads that have less than basic education. 4 Reference category are household heads that report not being in the labor force, other than the retired. Reference category is households located in the urban North. * Significant at 0.10 level ** Significant at 0.05 level *** Significant at 0.01 level Income As in other countries, households in non-Tirana Albania under-report their income. Households report income that is far lower than consumption. The differences are significant too large to be solely attributable to savings. This is one of the main reason that we use consumption rather than income as an indicator of household wealth. Assuming all income is proportionately under-reported, we can get some idea about how different households survive by looking -at the proportion of income received from different sources: wages, remittances, public transfers. (Fig l.x) Cash transfers are the most important source of income for households. Pensions, the Ndihme Ekonomike (NE), the social assistance transfer, followed by other public transfers and unemployment benefits, taken together comprise almost. Wages are the next most important source of income, 30%. Remittances comprise only 5 percent of total household income'. Figure A2. 1: Distribution of income by source, Albania 1996 Albai.ia Rent income Albania Other transfers 0% 9% Wages NE 31% Unemployment benefits 4% Non agriculture Pensions\ 007:0 tE :l:0\ \ \ 28% Remittances 6% The questionnaire does not provide information on agricultural income. 60 Figure A2. 2: Distribution of Income by Source of Income. RU1stIhr ~RentU Pensions IJnemploy Rur e Urbanc e 23% ment transfers benefits 10%= _ Wages temittanc 6% 1 0 ages ~~~~~~~~~~~es NE 24_24 % 5% \ \ 4 19% 4E _ _sl~~ ~ ~~~~~~on __ 9% Unempcu cture Non Unemp O% agriculture men 11% g'mther benefits -rasfers 2% Remittanc Wages r Rent 7% tanc Wages~ ~ ~ ~~~inom Pensions es 38% mcome 32% 6% 1% Table A2. 8: Average Area of Land Owned by Type Given in Dynyms (Number of Households Based On) in Albania 1996. Orchards Pastures Crop Land Area Owned by Farner Rural 0.883 (417) 3.252 (110) 6.094 (928) Urban 3.125 (4) 2.000 (1) 12.6133 (22) North 0.600 (231) 4.393 (25) 3.820 (543) South 1.124 (190) 2.983 (86) 8.010 (407) Total 0.945 (421) 3.221 (111) 6.439 (950) Area Owned by State Rural 1.142 (4) 0 7.975 (10) Urban 0 (0) 0 12.000 (1) North 0 (0) 0 8.000 (1) South 1.142 (4) 0 8.399 (10) Total 1.142 (4) 0 8.393 (11) Area Rented Rural 0.932 (2) 0 7.082 (9) Urban 0 (0) 0 0 North 0.200 (1) 0 2.458 (5) South 1.000 (1) 0 7.668 (4) Total 0.932 (2) 0 7.082 (9) 61 Table A2. 9: Average Area of Land Owned by Type Given in Dynyms (Number of Households Based On) in Albania 1996. Quintile 1 2 3 4 5 Area Owned by Farmer Orchards 0.712 0.942 0.798 1.324 1.240 Pastures 3.570 3.112 3.081 3.291 3.230 Crop land 4.748 5.627 7.176 7.743 7.096 Total 5.138 6.130 7.784 8.292 7.464 Area Owned by State Orchards 0.100 1.500 000 2.000 1.000 Pastures 0 0 0 0 0 Crop land 5.386 6.915 10.746 4.000 11.606 Total 5.409 7.670 10.746 3.470 6.883 Area Rented Orchards 0.200 1.000 0 0 0 Pastures 0 0 0 0 0 Crop land 3.610 5.543 10.000 10.000 0 Total 3.625 6.442 10 10 0 Table A2. 10: Average Land Area and Value in Rural Areas. Rural North Rural South Average area of land' 4.02 8.41 Number of households 574 410 Average value of land' 260,049 682,329 Number of households 574 410 Average value of crop land 246,047 632,596 Number of households 545 397 'The average area of land is calculated based on those that report owning, using state owned land or renting land for agricultural activity. It is calculated per household and the unit of measurement is given in dynyms. 2 average value of land is calculated for those that report owning land and is calculated per household in lek. 62 Table A2. 11: Average Value of Land Owned by Type for Households that Owns Each Type of Land in Albania 1996. Quintile 1 2 3 4 5 Owned by Farmer (per household) Orchards 26,740 74,381 69,222 99,637 161,432 Pastures 289,366 149,885 226,086 181,008 230,977 Crop land 220,331 580,965 610,825 666,029 720,198 Total 242,292 598,342 658,123 696,281 758,016 Owned by Farmer (per dynym) Orchards 51,420 82,354 94,570 101,731 152,341 Pastures 64,501 42,598 65,230 64,924 79,406 Crop land 47,470 88,681 85,396 98,020 98,683 Total 47,625 88,198 82,814 98,374 106,217 Rent Paid Last Year (per household) Orchards 500 0 0 0 0 Pastures 0 0 0 0 0 Crop land 7,043 1,015 20,000 0 0 Total 7,082 1,015 20,000 0 0 Note: Value is not available for land owned by the state. Figure A2. 3: Per Capita Consumption Density by Region, Albania 1996. 0.35 0.3 -)--North Q 0.25Sot 0.2 U 0.2 ' u 0.1 0.05 0 Monthly Per Capita Consumiption (Lek) 63 Figure A2. 4: Per Capita Consumption Distribution by the Age of the Household Head. 0.4 - 0.3 I*---- 16 to 35 years ---36 to 64 years d-|| --L 64 plus years 0.2 2 0 0.1 0 Per Capita Consumption per Month (Lek) Figure A2. 5: Per Capita Consumption Density by the Employment Status of Household Head. 0.4 - ---)K -- Wage employed - - - - Agriculture (non-wage) _ 0.3 Unemployed ---*-->< Retired ~0.2 0.1 N O* IIII r~ =-*w*:4---lC Per Capita Consumption per Month (Lek) 64 Figure A2. 6: Cumulative Distribution of Consumption in Urban and Rural Areas,. Albania 1996. o Urban a Rural CL 0 E~ .2 0 0 5000 10000 15000 200'00 250'00 300'00 350'00 PC Consumption (in Lek) Figure A2. 7: Cumulative Distribution of Consumption in the North and South, Albania 1996. o North A South 0 .15~~~~~~6 0 .4 E 2 C.) o 5000 10000 15000 20000 25000 30000 35000 PC Consumption (in Lek) Fiur A.7:CuuatveDitibtinofCosupio i heNothad6ouh Table A3. 1: The Composition of the Population, Active Labor Force, Employment and Unemployment, Albania Population Working Age Labor Force Employment Unemployment Not in Labor Absent Long term Population Activity Force Unemployment TOTAL 100.0 100 100 100.0 100.0 100.0 100.0 100.0 Number l(n) 4816 4610 3198 2537 661 881 278 464 Gender Male 50.7 51.27 54.78 56.5 49.5 32.4 85.1 48.3 Female 49.3 48.73 45.22 43.5 50.5 67.6 14.9 51.7 Age 15-19 13.8 15.21 8.06 7.0 11.8 14.6 11.5 7.2 20-29 25.3 27.84 27.73 23.8 40.1 14.2 63.8 42.4 30-39 22.1 24.24 28.60 28.6 28.7 9.4 19.0 30.9 40-49 15.6 17.14 20.37 22.7 15.2 7.3 3.8 15.7 50-59 9.9 10.92 9.93 11.7 4.2 13.8 0.9 3.7 60+ 13.3 4.62 4.81 6.3 0.0 40.8 1.0 0.0 Education Less than basic 7.0 2.21 2.33 0.8 1.1 13.8 0 0.8 Basic 55.5 57.47 56.95 54.4 50.2 47.5 56.8 54.4 Secondary 32.7 35.10 35.00 41.7 44.9 34.7 39.4 41.7 Tertiary 4.9 5.22 5.71 3.1 3.9 4.1 3.8 3.1 Residence Urban 36.8 63.20 34.18 22.4 71.0 45.1 27.6 71.3 Rural 63.2 36.80 65.82 77.6 29.0 54.9 72.4 28.7 Region North 34.1 34.72 32.80 33.2 31.6 39.1 26.5 31.0 South 65.9 65.28 67.20 66.8 68.4 60.9 73.5 69.0 ' This is the un-weighted number of individuals in the sample and not the population. Population is defined to be all individuals aged 15 years and above and the working age population is all individuals aged 15-64 years. 66 Table A3. 2: Socio-Demographic Characteristics of Labor Force Participation Rate, Unemployment Rate and Long Term Unemployment Rate. Labor Force Employment Unemployment Long Term Participation Rate Rate Unemployment Rate Total 65.13 49.36 24.22 17.50 Age 15-19 37.93 24.45 35.55 7.02 20-24 69.64 44.69 35.81 22.99 25-29 73.24 48.23 34.15 29.97 30-34 81.71 60.95 25.41 21.86 35-39 87.44 67.23 23.12 19.41 40-44 86.96 70.29 19.17 16.18 45-49 87.48 73.54 15.94 12.34 50-54 72.81 62.42 14.26 9.93 55+ 31.73 31.16 1.79 1.26 Gender Male 70.34 54.93 21.90 15.23 Female 59.77 43.61 27.04 20.25 Place of Residence Rural 67.81 60.57 10.68 7.29 Urban 60.52 30.08 50.30 37.16 Region North Urban 57.62 29.87 48.15 30.60 NorthRural 64.63 54.86 15.11 11.58 South Urban 61.50 30.14 50.98 39.23 South Rural 69.87 64.25 8.03 4.72 Education LessthanBasic 21.84 19.44 11.01 6.14 Basic 66.79 52.54 21.33 16.73 Secondary 69.79 48.09 31.08 20.83 University 76.79 64.22 16.37 9.41 67 Table A3. 3: Socio-Demographic Composition of the Unemployed: Never Worked (NWU), Short-term (STU), and Long-term (LTU) Unemployed. NWU STU LTU Age 16-19 59.51 16.35 3.23 20-24 37.00 18.84 18.80 25-29 3.49 12.04 22.99 30-34 0 13.22 17.92 35-39 0 13.72 15.81 40-44 0 8.66 10.30 45-49 0 9.11 6.87 50-54 0 6.86 3.46 55+ 0 1.20 0.62 Gender Male 55.33 52.78 47.67 Female 44.67 47.22 52.33 Place of Residence Rural 45.11 25.15 27.41 Urban 54.89 74.85 72.59 Region North Urban 21.12 21.96 14.32 North Rural 16.94 10.21 16.30 South Urban 33.77 52.89 58.27 South Rural 28.16 14.94 11.12 Education Illiterate 0 1.86 0.82 Literate 0 1.10 0 Elementary 1.38 2.95 3.78 Middle 34.85 38.85 50.66 Secondary 28.23 21.26 16.71 Technical 28.64 28.68 24.97 Post Vocational 0 0 0.44 University 6.90 5.31 2.63 68 Table A3. 4: Socio-Demographic Composition Public and Private Sector Employment. Government State Private Total Total 100 100 100 100 Age 16-19 0.15 2.20 4.53 2.15 20-24 4.75 3.39 16.60 7.39 25-29 15.02 5.00 9.25 9.48 30-34 9.87 14.54 19.71 14.37 35-39 17.24 23.97 14.03 19.06 40-44 20.71 18.46 12.56 17.63 45-49 23.00 17.44 13.34 18.19 50-54 5.35 7.78 7.17 6.81 55+ 3.90 7.22 2.80 4.93 Gender Male 48.21 73.03 65.70 62.79 Female 51.79 26.97 34.30 37.21 Place Rural 40.51 37.41 42.66 39.85 Urban 59.49 62.59 57.34 60.15 Region North Urban 20.93 14.06 11.59 15.68 North Rural 15.72 18.61 7.44 14.65 South Urban 38.56 48.53 45.75 44.46 South Rural 24.79 18.80 35.22 25.21 Education Basic 14.54 34.40 47.30 31.25 Secondary 50.44 50.76 48.53 50.05 University 35.02 14.84 4.18 18.70 69 Table A3. 5: The Profile of New Hires (as Defined by Elections), Albania. Hired after elections Yes No National Government State Private TOTAL 100.0 100.0 100.0 100.0 100.0 Gender Male 45.4 65.1 49.9 72.9 66.0 Female 54.6 34.9 50.1 27.1 34.0 Age 15-24 17.6 40.4 18.1 26.0 23.0 25-34 25.5 21.6 40.3 10.7 29.7 35-44 20.4 18.3 26.1 21.1 26.2 45-54 13.1 13.7 15.5 39.3 20.2 55+ 23.4 6.0 0.0 2.9 0.9 Education Basic 24.5 41.6 16.1 35.4 47.7 Secondary 52.0 45.8 51.8 43.8 46.6 Tertiary 23.5 12.6 32.1 20.8 5.7 Residence Rural 33.9 43.5 52.4 55.9 57.0 Urban 66.1 56.5 47.6 44.1 43.0 Region North 34.2 33.8 41.3 38.1 18.9 South 65.8 66.2 58.7 61.9 81.1 Sector Government 41.5 20.3 -- -- -- State 51.3 21.7 -- -- -- Private 7.1 58.0 -- -- -- Industry Manufacturing 22.7 20.4 1.8 14.6 29.1 Construction 4.8 19.1 1.7 15.4 26.7 Science/Education 21.5 5.9 20.2 8.2 0.0 Health Care 11.4 4.7 13.3 9.2 0.0 Army/Police 8.8 6.7 26.4 0.0 2.4 Others 30.8 43.2 36.6 52.6 41.8 Note: Data for the national economy are not directly comparable with the data by public/private sector since the latter were calculated using a subsample of workers with known sector affiliation New hires=workers who have held their current job for three years or less. All those in wage employment have achieved at least a basic level of education. 70 Table A3. 6: The Profile of New Hires (as Defined by Tenure), Albania. Tenure with the firm More than Three years or less three years National Government State Private TOTAL 100.0 100.0 100.0 100.0 100.0 Gender Male 55.4 66.4 48.9 70.4 65.5 Female 44.6 33.6 51.1 29.6 34.5 Age 15-24 12.7 37.5 16.1 29.3 23.4 25-34 27.7 21.0 39.4 9.1 30.1 35-44 29.4 21.1 27.3 17.8 26.0 45-54 18.2 17.7 17.2 43.2 20.5 55+ 12.0 2.7 0.0 0.7 0.0 Education Basic 25.4 43.1 11.0 54.4 48.2 Secondary 52.9 46.4 58.0 33.7 47.4 Tertiary 21.7 10.4 31.0 11.9 4.4 Residence Urban 17.0 39.3 54.3 56.1 57.2 Rural 83.0 60.7 45.7 43.9 42.8 Region North 34.7 33.5 42.8 37.1 18.6 South 65.3 66.5 57.2 62.9 81.4 Sector Government 40.1 19.2 -- -- -- State 52.9 20.4 -- -- -- Private 7.0 60.4 -- -- -- Industry Manufacturing 23.2 21.0 2.0 16.4 28.7 Construction 5.5 18.9 0.0 14.6 26.4 Science/Education 19.9 5.9 21.8 8.3 0.0 Health Care 11.5 5.0 14.9 10.4 0.0 Army/Police 8.9 6.6 27.0 0.0 2.4 Others 31.0 42.6 34.3 50.3 42.5 Note: Data for the national economy are not directly comparable with the data by public/private sector since the latter were calculated using a Subsample of workers with known sector affiliation New hires=workers who have held their current job for three years or less. 71 Table A3. 7: Estimates of Human Capital Earnings Functions (OLS) - Albania. Independent variables Dependent variable: log net monthly wages (lek) All workers Public Sector Rural Urban (1) (2) (2) (3) Intercept 8.56* 8.56* 8.64* 8.54* Education Dummtes: Secondary 0.25* 0.30* 0.16* 0.31* Tertiary 0.35* 0.41* 0.15 0.44* Female -0.22* -0.17* -0.17** -0.19* Experience' 0.002* 0.002* 0.002 0.002* Square(Experience)/ 100 -0.005* -0.005* -0.006** -0.005 Private sector 0.24* 0.19* 0.31 Urban residence 0.06 0.00 Significance tests of equality in returns between levels of education (p-values) 2 Compare Secondary with Tertiary 0.08 0.04 0.96 0.11 No. of observations 493 389 239 254 F-statistic 4.93 5.50 2.44 3.97 R-Squared 0.1877 0.2395 0.1908 0.2542 Adj. R-Squared 0.1496 0.1960 0.1124 0.1902 Root MSE 0.453 0.375 0.424 0.467 *Significant at 5 percent level. **Significant at 10 percent level. 'At a current job. 2 A significant p value indicates that the returns from the two levels of education being compared is significantly different and that from the higher level is higher. For example in the first column the p value of 0.08 means that the probability that wage returns from secondary and tertiary education is equal is 0.08. In other words wage returns to tertiary level education is significantly higher than that from only secondary level education with a p value of 0.1. 72 Table A3. 8: Average and Median Net Monthly Wage, Albania. Average Coefficient of Median Variation 1 TOTAL 7,682.2 17.47 7,000 Gender Male 8,070.9 13.13 7,200 Female 7,019.1 24.36 6,000 Age 15-24 9,742.9 32.95 6,000 25-34 6,965.6 12.31 6,400 35-44 7,757.8 12.65 7,000 45-54 7,521.2 13.53 7,000 55+ 7,435.7 13.52 7,000 Education Basic 6,623.4 15.90 6,000 Secondary 8,360.4 19.65 7,000 Tertiary 7,634.3 9.44 7,700 Residence Urban 8,045.3 21.76 7,000 Rural 7,133.7 9.17 6,600 Region North 7,176.9 9.96 6,850 South 7,906.4 22.13 7,000 Sector Government 6,434.8 8.82 6,650 State 7,527.8 9.13 7,000 Private 9,445.5 27.68 7,500 Industry Manufacturing 9,187.5 25.82 7,190 Construction 7,739.5 11.72 6,500 Agriculture 6,976.0 12.25 7,100 Forestry 7,176.5 6.56 9,000 Transport 13,116.4 20.62 8,500 Communication 9,797.7 17.08 7,200 Trade 6,989.4 7.88 6,000 Commercial Services 5,366.4 7.74 5,000 Other Production Activities 8,478.7 16.30 7,150 Science/Education 6,060.3 8.86 7,000 Arts/Culture 4,779.4 13.01 6,000 Health Care 6,247.9 7.78 6,000 Finance/Credit 8,365.6 9.88 7,000 Management/Administration 8,179.0 5.48 8,000 Army/Police 7,300.0 7.67 7,750 Others 7,017.5 8.47 6,000 'Coefficient of variation is defined to be =standard deviation/mean. 73 Table A3. 9: Sector of Employment by Industry, Albania. Government State Private National Industry Manufacturing 1.1 36.4 25.8 21.8 Construction 0.4 5.4 30.1 10.4 Agriculture 2.7 3.3 0.0 2.2 Forestry 0.0 0.9 1.7 0.8 Transport 0.5 2.7 6.9 3.1 Communication 3.5 7.1 2.0 4.5 Trade 0.0 0.2 7.6 2.2 Commercial Services 0.0 1.9 7.0 2.6 Other Production Activities 0.0 0.9 7.0 2.2 Science/Education 36.5 8.3 0.0 15.4 Arts/Culture 2.9 1.2 0.7 1.6 Health Care 17.5 6.6 1.3 8.8 Sport/Tourism 0.7 0.0 0.0 0.2 Finance/Credit 3.3 4.0 1.4 3.0 Management/Administration 6.9 2.5 0.0 3.3 Army/Police 18.4 3.4 2.0 8.0 Others 5.7 15.6 6.5 9.9 TOTAL 100.0 100.0 100.0 100.0 Table A3. 10: Sector of Employment by Industry, Albania. Government State Private Total Industry Manufacturing 1.7 66.3 32.1 100.0 Construction 1.3 20.6 78.1 100.0 Agriculture 40.9 59.1 0.0 100.0 Forestry 0.0 42.7 57.3 100.0 Transport 5.4 34.1 60.5 100.0 Communication 25.9 61.9 12.2 100.0 Trade 0.0 4.2 95.9 100.0 Commercial Services 0.0 28.4 71.6 100.0 Other Production Activities 0.0 15.1 84.9 100.0 Science/Education 78.7 21.3 0.0 100.0 Arts/Culture 59.7 28.4 11.9 100.0 Health Care 66.2 29.7 4.2 100.0 Sport/Tourism 100.0 0.0 0.0 100.0 Finance/Credit 36.4 51.6 12.0 100.0 Management/Administration 70.3 29.7 0.0 100.0 Army/Police 76.6 16.7 6.7 100.0 Others 19.3 63.0 17.8 100.0 TOTAL 33.2 39.7 27.1 100.0 74 Table A3. 11: Type of Employment Contract by Industry, Albania. Termless Fixed Term No Contract Commission Other Industry Manufacturing 23.5 21.2 15.0 23.8 100.0 Construction 1.9 23.6 34.1 20.8 0.0 Agriculture 1.5 4.7 3.3 0.0 0.0 Forestry 0.7 0.0 1.9 0.0 0.0 Transport 2.3 2.9 6.4 0.0 0.0 Communication 5.6 2.2 2.2 0.0 0.0 Trade 0.4 4.1 7.8 0.0 0.0 Commercial Services 1.1 3.3 8.3 0.0 0.0 Other Production Activities 1.0 12.5 0.9 0.0 0.0 Science/Education 19.4 6.8 5.4 27.7 0.0 Arts/Culture 1.9 1.8 0.6 0.0 0.0 Health Care 12.0 0.0 2.2 0.0 0.0 Sport/Tourism 0.0 1.9 0.0 0.0 0.0 Finance/Credit 4.2 0.0 0.8 0.0 0.0 Management/Administration 3.7 3.7 1.7 0.0 0.0 Army/Police 10.5 0.0 2.6 27.7 0.0 Others 10.5 11.2 7.2 0.0 0.0 TOTAL 100.0 100.0 100.0 100.0 100.0 Table A3. 12: Type of Employment Contract by Industry, Albania. Termless Fixed Term No Contract Commission Other Total Industry Manufacturing 74.9 10.8 12.7 0.8 0.9 100.0 Construction 13.0 25.1 60.6 1.4 0.0 100.0 Agriculture 48.4 23.9 27.7 0.0 0.0 100.0 Forestry 56.6 0.0 43.4 0.0 0.0 100.0 Transport 51.5 10.4 38.1 0.0 0.0 100.0 Communication 85.5 5.5 9.0 0.0 0.0 100.0 Trade 11.9 21.2 66.9 0.0 0.0 100.0 Commercial Services 28.4 13.8 57.8 0.0 0.0 100.0 Other Production Activities 30.1 62.4 7.5 0.0 0.0 100.0 Science/Education 87.4 4.9 6.5 1.3 0.0 100.0 Arts/Culture 81.5 11.9 6.6 0.0 0.0 100.0 Health Care 95.4 0.0 4.6 0.0 0.0 100.0 Sport/Tourism 0.0 100.0 0.0 0.0 0.0 100.0 Finance/Credit 95.3 0.0 4.7 0.0 0.0 100.0 Management/Administration 78.1 12.6 9.3 0.0 0.0 100.0 Army/Police 91.5 0.0 6.0 2.4 0.0 100.0 Others 74.0 12.5 13.5 0.0 0.0 100.0 Public 88.0 6.5 4.8 0.7 0.0 100.0 Private 19.7 23.4 55.5 0.62 0.73 100.0 TOTAL 69.5 11.1 18.5 0.7 0.2 100.0 75 Table A3. 13: Sector and Length of Employment Contract by Industry, Albania Length of Employment 1 Year or 1-3 Years More then 3 Average Less Years Industry Manufacturing 14.38 30.64 23.16 9.9 Construction 22.78 13.36 5.50 3.1 Agriculture 2.77 0 2.61 9.5 Forestry 2.15 0 0.58 3.4 Transport 4.61 6.87 1.72 6.7 Communication 3.34 3.00 5.51 9.0 Trade 6.89 2.57 0.42 1.5 Commercial Services 6.72 3.13 1.16 4.0 Other Production Activities 3.77 3.50 0.83 4.8 Science/Education 4.63 7.70 19.90 15.6 Arts/Culture 1.37 0 2.18 14.1 Health Care 4.12 6.20 11.45 13.9 Sport/Tourism 0.98 0 0 0.1 Finance/Credit 2.93 0.87 2.92 10.6 Management/Administration 0.85 6.17 3.57 11.4 Army/Police 5.46 8.33 8.85 9.5 Others 12.26 7.67 9.61 9.2 TOTAL 100 100 100 9.7 Public 38.0 43.0 92.5 12.7 Private 62.0 57.0 7.5 2.0 Public 11.2 9.0 79.8 100 Private 49.9 32.5 17.6 100 Total 20.0 14.4 65.6 100 Table A3. 14: Summary of Earnings Distribution by Sector, Albania. National Government State Private P5 2,000 0 4,000 1,250 PIO 3,700 2,250 4,000 3,000 P25 5,000 5,000 5,400 4,580 P50 7,000 6,650 7,000 7,500 P75 8,500 8,000 8,400 10,000 P90 12,000 9,400 12,000 16,000 P95 14,500 12,000 14,000 21,000 Gini coefficient 0.303 0.246 0.231 0.417 76 Table A3. 15: The Incidence and Composition of Low-Paid Employment, Albania Low paid Employment (< 2/3 median Wage) Composition (%) Incidence(%) All Workers 18.4 Gender Male 55.4 16.1 Female 44.6 22.1 Age 15-24 13.0 25.0 25-34 29.6 22.7 35-44 28.5 14.3 45-54 25.0 18.3 55+ 4.0 14.9 Education Basic 47.3 27.8 Secondary 36.9 13.5 Tertiary 15.8 15.5 Residence Urban 39.6 18.2 Rural 60.4 18.4 Region North 27.4 16.4 South 72.6 19.2 Sector Government 33.4 18.5 State 30.1 13.9 Private 36.5 24.8 Industry Manufacturing 21.3 17.9 Construction 9.3 16.4 Science/Education 20.0 23.8 Health Care 6.2 13.0 Army/Police 4.1 9.3 Others 39.2 20.2 77 Table A3. 16: Contribution of Selected Variables to Log-Earnings Inequality. In % of: Variable Total Variance Explained Variance Education 5.02 15.89 Female 3.36 10.64 Experience 0.02 0.07 Private sector 3.09 9.77 Urban 0.34 1.07 Sector 19.75 62.56 Total explained (R2) 31.57 100.00 Unexplained 68.43 Total 100 Note: The contribution of an variable x to the variance of the log-earnings w was calculated as b*r (w, x), where b is the standardized regression coefficient, and r is the correlation coefficient. The contribution is negative when the regression coefficient and the correlation coefficient differ in sign. For example, the correlation between the middle level of education variable and earnings is negative while the impact of the middle level of education on earnings after controlling for the impact of other variables is positive. The contribution of a categorical variable (e.g. education) is measured as a sum of contributions by binary regressors representing each category (e.g. primary education, secondary education). The contribution of a single binary regressor (e.g. tertiary education) can be greater than the contribution of the categorical variable as a whole (e.g. education) if the contribution of some other binary regressors (e.g. primary education) is negative. 78 Table A4. 1: Gross Enrollment Rates by Level of Education, 1990 to 1997 1990 1993 1997 Primary Male 101.5 99.1 97.7 Female 102.9 99.7 96.5 Total 102.1 99.4 97.1 Lower Secondary Male 103.7 89.0 89.3 Female 101.1 92.4 92.1 Total 102.5 90.7 91.1 Upper Secondary In general 26.0 30.1 33.9 In vocational 53.1 12.3 6.4 Total 78.0 42.4 40.3 Tertiary Total 9.0 11.7 13.6 Gross enrollment rate equals the total number of children enrolled at a given level/number of children in that age group. Data are for all of Albania. Source: Statistical Office of Albania. Table A4. 2: Public Expenditures on Basic Education by Gender and Expenditure Quintile Group Male Female Total Total per capita Per capita Percent of Per capita Percent of Per capita Percent of expenditure expenditure total expenditure total expenditure total quintile (lek) expenditure (lek) expenditure (lek) expenditure Lowest 4,882 23.6 5,562 30.7 5,230 27.0 Second 5,842 18.2 5,063 19.5 5,429 18.8 Third 5,306 22.9 5,392 20.7 5,345 21.8 Fourth 5,213 21.3 6,116 19.6 5,591 20.5 Top 5,624 14.0 4,705 9.5 5,231 11.8 Total 5,363 100.0 5,308 100.0 5,425 100.0 79 Table A4. 3: Public Expenditures on Secondary Education by Gender and Expenditure Quintile Group Male Female Total Total per capita Per capita Percent of Per capita Percent of Per capita Percent of expenditure expenditure total expenditure total expenditure total quintile (lek) expenditure (lek) expenditure (lek) expenditure Lowest 2,762 8.7 2,101 5.9 2,423 7.2 Second 5,698 19.9 10,993 22.3 7,855 21.2 Third 6,416 14.8 13,456 33.1 10,342 24.7 Fourth 6,843 17.6 6,024 12.4 6,443 14.8 Top 13,813 39.1 14,112 26.3 13,944 32.2 Total 6,972 100.0 8,918 100.0 7,907 100.0 Table A4. 4: Public Expenditures on Tertiary Education by Gender and Expenditure Quintile Group Male Female Total Total per Per capita Percent of Per capita Percent of Per capita Percent of capita expenditure total expenditure total expenditure total expenditure (lek) expenditure (lek) expenditure (lek) expenditure quintile Lowest 1,631 5.1 3,116 9.8 2,394 7.5 Second 3,668 16.5 3,352 15.5 3,503 16.0 Third 2,102 9.8 11,249 42.7 6,324 26.7 Fourth 4,834 21.8 3,996 14.9 4,445 18.2 Top 9,343 46.8 4,452 17.1 7,157 31.6 Total 4,584 100.0 5,227 100.0 4,892 100.0 80 Table A4. 5: Net, Gross and Age-Specific Enrollment Rates by Educational Level and Expenditure Quintile Group. Basic education Secondary education Tertiary education Total per capita Net Age- Gross Net Age- Gross Net Age- Gross quintile enrollment specific Enrollment enrollment specific Enrollment enrollment specific Enrollment enrollment enrollment enrollment Lowest 79 80 94 6 43 10 2 7 2 Second 79 79 97 27 58 34 2 9 2 Third 81 82 96 32 69 44 4 12 4 Fourth 83 83 100 21 63 28 3 7 3 Top 73 79 94 47 78 60 5 12 5 Total 79 81 96 25 61 34 3 10 3 Table A4. 6: Distribution of Public Transfers by Expenditure Quintile Group. Total per Average total Transfers as percent of total per capita capita per capita monthly expenditures Expenditure monthly Ndihme Unemployment Social Disability Old age Other Quintile expenditures social benefits welfare allowance pensions pensions (lek) assistance payments Lowest 2260.0 4.6 1.1 0.7 0.1 3.7 0.8 Second 3362.6 2.4 0.8 1.1 0.3 5.1 1.3 Third 4380.5 1.4 0.5 0.7 0.1 6.8 0.9 Fourth 5778.5 0.9 0.4 0.4 0.1 5.1 0.5 Top 9251.6 0.2 0.3 0.2 0.1 7.4 1.0 Total 5431.9 1.1 0.5 0.4 0.1 6.3 0.9 81 Table A4. 7: Age specific, Gross and Net Enrollment Composition of 6 to 24 Year Olds, Albania 1996, | Educational Age Groups | 6-13yrs 14-17yrs 18-24yrs | Total ----------+-_----------------__________------_+--------- Basic | 365,143 75,824 1,738 | 442,705 1 82.48 17.13 0.39 1 100.00 1 79.43 35.41 0.54 1 44.44 ----------+_________________________________+-_________ Secondary I 0 54,586 18,155 1 72,741 1 0.00 75.04 24.96 1 100.00 I 0.00 25.49 5.63 1 7.30 ___________+_________________________________+__________ Tertiary I 0 0 11,191 | 11,191 I 0.00 0.00 100.00 I 100.00 I 0.00 0.00 3.47 1 1.12 ----------+_________________________________+…_________ Not in | 94,561 83,696 291,246 | 469,503 School | 20.14 17.83 62.03 | 100.00 1 20.57 39.09 90.36 1 47.13 - - - - - - - - --+- _ _ _ _ _ _ _ _ _ _ _ _ _ _ + …__ _ _ _ _ _ _ _ Total 459,704 214,106 322,330 | 996,140 46.15 21.49 32.36 1 100.00 100.00 100.00 100.00 I 100.00 82 Table A4. 8: Old Age Pensions: Households Receiving Benefits. Per Capita Urban Rural Expenditure Percent of households in Percent of households in Decile decile receiving old age decile receiving old age pensions pensions 1 10.5 22.5 2 10.3 29.6 3 24.3 31.1 4 35.1 33.3 5 37.7 47.7 6 30.9 39.5 7 26.0 25.4 8 35.2 32.8 9 36.4 38.7 10 45.8 43.8 Total 34.5 34.1 Table A4. 9: Percent of Total Per Capita Assistance from Old Age Pensions Received by Decile Urban Rural Per Capita Percent of total per capita Percent of total per capita Expenditure assistance received by assistance received by decile decile decile 1 0.2 4.0 2 0.4 4.8 3 2.2 5.2 4 5.0 6.0 5 4.9 14.2 6 7.1 11.3 7 9.1 5.4 8 12.1 10.1 9 18.1 17.0 10 40.8 22.1 Total 100.0 100.0 83 Distributors of World Bank Group Publications Prices and credet terms vary from CZECH REPUBLIC INDIA Eulyoo Publshing Co, Ltd. PERU SWEDEN countr to country ConsLft your USIS, NIS Prodejna Allied Publishers Ltd. 46-1, Susong-Dong Editorial Desarrollo SA Wennergren-Williams AB local distributor before placing an Havelkova 22 751 Mount Road Jongro-Gu Aparado 3824, Ica 242 OF 106 P0. Boo 1305 order 130 00 Prague 3 Madras - 600 002 Seoul Lma 1 5-1 71 25 Solna Tel: (420 2) 24231486 Tel: (91 44) 852-3936 Tel: (82 2) 734-3515 Tel: (51 14) 285380 Tel: (46 8) 705-97-50 ARGENTINA Fax: (420 2) 24231114 Fax: (91 44) 852-0649 Fax: (82 2) 732-9154 Fax: (51 14) 286628 Fax: (46 8) 27-00-71 World Publications SA URL: httpI/www.nis.cz/ INDONESIA LEBANON PHILIPPINES E-mail: mailWwwi.se Ac. Cordoba 1677 Cdd de B o A DENMARK Pt. Indira Limited Librairie du Liban International Booksource Center Inc. SWITZERLAND Tel: (54a11) 4815-156 SamfundsLitteratur Jalan Borobudur 20 PO. Box 11-9232 1127-A Antipolo St, Barangay, Librairie Payot Service Instiutionnel Fan: (54 11) 4815-8156 RosenoemsAllet 11 PO. Box 181 Beirut Venezuela C(tm)tes-de-Montbenon 30 E-mail: wpbooks@infovia.com.ar DK-1970 Frederiksberg C Jakarta 10320 Tel: (961 9) 21 7 944 Mukati City 1002 Lausanne Tel: (45 35) 351942 Tel: (62 21) 390-4290 Fan: (961 9) 217 434 Tel: (63 2) 896 6501: 6505; 6507 Tel: (41 21) 341-3229 AUSTRALIA, FIJI, PAPUA NEW Fax: (45 35) 357822 Fax: (62 21) 390-4289 E-mail: hsayegh@librairie-du- Fax: (63 2) 8961741 Fax: (41 21) 341-3235 GUINEA, SOLOMON ISLANDS, URL: http:/lwww.sl.cbs.dk IRAN liban.com lb VANUATU, AND SAMOA ECUADOR Ktab araCo Pabisfers URL: ht1p://ww.librairie-du- POLAND ADECO Van 4 D.A. Information Services ECibOunta craCoomblshrsInternational Publishing Service EdaionsTechniques 648 Whitehorso Rnad Libri Mundi Khaled Eslambali Ave., 6tb Street lia.o.bUl. Piekna 31137 Ch. do Lacuez 41 M ichum 3132. Victnria Libreria Internacional Delafrooz Alley No. 8 MALAYSIA 00-677 Warzawa CH1807 Blonav Tel: (61)3 9210 7777 Po. Box 17-01-3029 PO. Box 15745-733 Universityof Malaya Cooperative Tel: (48 2) 628-6089 Tel: (41 21) 943 2673 Fax: (61) 3 9210 7788 Juan Leon Mera 851 Tehran 15117 Bookshop, Limited Fax: (48 2) 621-7255 Fax: (41 21) 943 3605 E-mail: servico@dadirect.com as Quito Tel: (98 21) 8717819, 8716104 PO. Box 1127 E-mail: books%ipsQikp.atm.com.pl THAILAND URL: http:/wwwdadirect.com.au Tel: (5932) 521-606: (5932)544- Fanx: (9 21) 8712479 Jalan Pantai Baru URL: Central Books Distribution 185 E-mail: ketath-saraia'edla.netir 59700 Koala Lumgor http:1/www.ipscg.waw.pl/ips/export 306 Silom Road AUSTRIA EaibruFan: (5932)504-209 Kowkab Publishers Tel: (60 3) 756-50 PORTUGAL Bankok 10500 Gerald and Co E-mail: librimul@libirimandli.com .ec P0 Boo 19575-511 Fan: (60 3) 755-4424 Livraria Portogal Tel: 166 2)2336930-9 Weihburgansse 26 E-mail: librimu2@mlibrimuncli.com .ec Tehran E-mail: umkoopwtm.net.my Apartada 2681. Rua Do Cane Fan: (66 2) 237-8321 A-1011 24-e 1 CODEU Tel: (98 21) 258-3723 MEXICO o 70-74 TRINIDAD & TOBAGO Tel: (431)512-47-31-20 Ruiz deCastilla 763, Edif Expocolor Fax: (98 21) 258-3723 INFOTEC 1200 Lisbon AND THE CARRIBBEAN URL: htp:7/ Pgerold.co/at.online Primer piso, Of #2 IREND Av. San Fernando No. 37 Fa: (1) 347-0264 Systematics Studies Ltd. BANGLADESH TeVFax. (593 2) 507-383; 253-091 Government Supplies Agency 14050 Mexico, DF. St. Augustie Shoppig Censor Micro Industries Development E-mail: codeueimpsat.net.ec Oihg an tSolathirel (52 S) 624-2800 Ead M Toad, St Indie Houistne ScRoad 1 MIAS EGYPT, ARAB REPUBLIC OF Dublin 2 Fan: (52 5) 624-2822 ROMANIA Trinda &88 Tobgo.Wes3Inie hanmoedi lOArea Al Abram DistributioL Agency Tel: (3531)661-3111 E-mail: infotecitrtn.net.mx Compani De Librarii Bucuresti SA. Tel: (868) 645-8466 Dhakand 1209 e AlGUaRtLt a:(331 4527 httpi/r*tn.net.mo Str. Lipscani no. 26. sector 3 E-m'ail: tobein'rinidlad.net Tel:a (8802) 3242 Cairo ISRet aEL(53 7527 Mondi-Prensa Monica) S.A. do C V. Bucharest Fan: (880 2) 811188 Tel: (20 2) 578-6083 Yozmot Literature Ltd. c/Rio Panuco. 141-Colonia Fax: (40 1) 312 4000 UGANDA BELGIUM ~ ~ ~ ~~ ~ax 202 58683P.O. Ban 56055 CuutmcPD Boo 9997, Madhvani Building BeLGIUMnThe Middle East Observer 3 Yohanan Hasandlar Street 06500 Mexico, D F. RUSSIAN FEDERATION Plot 16/4 Jinja Rd. Jean Do Lannoy 41, Sherif Street Tel Aviv 61560 Tel: (52 5) 533-5658 Isdtatelstvo .cVes Mir> Kampla Av. du Roi 202 Cairo e Fax: (52 5) 514-6799 9a, Kolpachniv Pereulok Kmpala 1000 Brussels CaT &(972 3) 5285-397 (55546799.opscniy Peeao0Tl(5412116 Tel: (322) 538-5169 Tel: (202)393-9732 Fan: (9723) 5285-397 NEPAL Moco 17 (2564)8251468 Fax: (32 2) 538-0841 Fax: (20 2) 393-9732 R.O.Y Intemational Everest Media International Services Tel: (7 095) 917 874 9 E-mail: gus( swi5uganda com Fan: (322)538-0841 FINLAND PD Boo 13056 ~~~~~~~~ ~~~~~~~(P.) Ltd. Fan: (7 095) 917 92 59E-algo@wfund.m c TNLAND PO Box 13056PO BOo 5443 ozimarin@glasnet.ru UNITED KINGDOM BRAZIL KGPhmanox SIGPOE4AIA,3HN Microinfo Ltd. PublicacPes Tecnicas Internacionais Akateeminen Kirjakauppa Tel Aviv 61130 Kathmandu SINGAPORE TAWAN, CHINA P. Bo 3, Ome Park, Aon, Rua Peinoto Gomido. 209 oIN 00101 Helsinki TeL: (9723)6480039 Tel: (9771) 416 026 MYANMAR; BRUNEI Hampshire GU34 2PG FIN-00101 Helinki Fax: (97 3) 648 6039Fan: (977 1) 224 431 Hemisphere Publication Services HaphrGU 2P 01409 Sao Paulo. SP Tel: (358 0) 121 4418 E-mail: royil@netvision.net.il 41 Kallanq Pudding Road #04-03 England Tel: (5511) 259-6644 ~~Fax: (358 0) 121-4435 UL tp/wwryn.oi NETHERLANDS Golden 6eel Building Tel: (44 1420) 68488 Tel: (55 11) 2596644 SE-mail: akatilause'stockmann.h alestinan Authort /Middle East De Lindeboom/lnternationale Sing vre 349316 ian: (44 14w ank8m98 Fax: (55 11) 258-6990 URL. htpWwwwakateemmen.com Palest nia g~~~~~~~~~~-mal: ban@miroifo.o.u F-aal: (5511) 258-6990 oRLhO Itdes Information ervices Publicaties b.v.- Tel: ( 5) 741-5166 UL tp/wwirif.0u URL: http://www.uol br FRANCE PO.B. 19502 Jerusalem PO. Box 202, 7480 AE Haaksbergen Fax: (65) 742-9356 URL: stp/wwsmicroifocouk CANADA Editions Eska: DBJ Tel: (972 2) 6271219 Tel: (31 53) 574-0004 E-mail: ashgate@asianconnect.com The Stationery Office Renouf Publishn Co Ltd, 48, roe Gay Lussac Fan: (972 2) 6271634 Fan: (31 53) 572-9296 SLOVENIA 51 Nine Elms Lane 5369 Canotek Road ~ ~ 5005Paris LIBERIA lindeboogworldonline.n odn W D 5369 Canotek Road e 70 ITALY, 972 2 hE-mai lin:/wwwworIdo nIi li GospOdarski vestnik Publishing kilondo SW8 SDR Ottawa. Ontario K1J gJ3 Tel (33-1) 55-42-73-08 Licosa Commissionaria Sansoni SPA deboo GrohppP Fax: (44 171) 873-8242 Tel: (613) 745-2665 Fan: (33-1) 43-29-91-67 Via Duca Di Calabria, 1/1 Dunajska cesta Fan: (44p171)w873-8242 Fax: (613) 745-7660 GERMANY Casella Postale 552 NEW ZEALAND 1000 L'ubjana URL: httpf//wwwthe-statiotnery E-mail: UNO-Verlag 50125 Firenze EBZCO UND Tel: (38661)1338347:1321230 afficecouk/ ordecdept@renoolbooks.com PoppelsdorferAllee 55 Tel: (39 55) 645-415 PriveEBSCO NZ Ltd. Fax: (386 61) 133 80 30 VENEZUELA URL: hdrp/ www.renouabooks.com 53115 eonn Fan: (39 55) 641-257 PrNate Marl Bag 99914 E-mail: repansekj@gvestnik.si Tecni-Ciencia Libros, S.A. CHINA Tel. (49 228) 949020 E-mail: licosa@nbccit Auckland SOUTH AFRICA, BOTSWANA Contro Cuidad Comercial Tamanco China Financial & Economic Fax:4922821: (21p: h/wp:wnwww.ftbccJticosa Tel: (64 9) 524-8119 For single titles: Tel (58 2) 959 5547: 5035: 0016 Publishing House E-mal: http:www.uno-veriag-d comJlanMRandeICA b Ud. Fax. (64 9) 524-8067 Oxford Universn.y Press Sou4hem FaU r (r8 2) 959 5637 5035: 00s Publ e ishing Houe E-mail: unoserlaguatoo cam Ian Randle Publishers Ltd. Africa Fn 5 )9953 Bejmng GHANA 206 Old Hope Road, Kingston 6 Oasis Official Vasco Boulevard. Goodwood ZAMBIA Tel: (86 10) 6401-7365 Epp Books Services Tel: 876-927-2085 PD. Box 3627 PO. Box 12119, Ni City 7463 University Bookshop, University of Fax: (8610) 6401-7365 PD. BoX 44 Fax: 876-977-0243 Wellington Cape Town Zambia China Book Import Centre TUC E-mail: irpl@colis.com Tel: (64 4) 4991551 Tel: (27 21) 595 4400 Great East Road Campus PO. Boo 2825 Accra JAPAN Fax: (64 4) 4991972 Fax: (27 21) 695 4430 PO. Box 32379 Tel: 223 21 778843 EAsten Book Service E-mail: oasisctactrix.gen.nz E-mail: oxford@oup.co.za Lusaka Beiling Eastern Book Service ~HonnaURL: http://wwwoasisbooks.co.nz/ For subscription orders: TeFa: (260 1)2539526 Chinese Corporation for Promotion GREECE Tokyo I c 8 URIA International Subscription Service of Humanities PaaoiiuSA e:(13 8806 niGerIAyPes iie PDO. Boo 41095 ZIMBABWE 52, You Fang Hu Tong, P aStouro SA. Tl: (81 3) 3818-0861 Aadcan Baobab Books (Pvt.) CuanrNera Do Jie2 35 Sooura Dilston Afaa:a (81k 3381-086 Univrsit Pres Limitedni-rn irs SA ax 6 4711 XuanNeiDado Aereo 3Three Crowns Building Jericho Criga, Academic and Biba Book AnCtl3 BeSi 10082 Athens E-mail: ordersnfsvt-ebs cojp Private Mail Bag 5095 Joh g 2802. 4 4Ltd. 6, G Tel: (86 10) 660 72 494 Tel: (3041364-1826 URL: 9badan Tel: (2711) 860-144 4 Coa R G Fa: (86 10) 66072494 Fax: (301) 364-8254 http://www.bekkoame.orjp/-svt- Tel: (234 22) 41-1356 Fan: (27 11) 88-64 PD5 B996 COLOMBIA HAITI ohs Fax: (234 22) 41-2056 URL 263 4 755035 Infoenlace Ltda. Cuetore Diffusion KENYA SPAIN Tel: 263 4 755135 Carrera 6 No. 51-21 5. Rue Capois Africa Book Service (E.A.) Ltd. PAKISTAN Mondi-Prensa Libras, S.A. Fn 64711 Apartado Aereo 34270 C.P) 257 Qaran House, Mfangano Street Mirza Book Agency Costello 37 Santafe do Bogota, D.C. Port-au-Prince PO. Box 45245 65, Shahrah-e-8 uaid-e-Azam 28001 Madrid Tel: (571) 285-2798 Tel: (509) 239260 Nairobi Lahore 54000 Tel: (34 91) 4 363700 Fan: (57 1) 285-2798 Fan: (509) 23 4858 Tel: (254 2) 223 641 Tel: (92 42) 735 3001 Fan: (34 91) 5 753998 COTE D'IVOIRE HONG KONG, CHINA. MACAO Fax: (254 2) 330 272 Fax: (92 42) 576 3714 E-mail: libreria @mundiprensa.es Asia 2000 Ltd. Le=a Books URL: ht1p://www.mundiprensa.com/ Center dEdit ion et de Diffusion 2000 Ltd. Oxford University Press Mundi-Prensa Barcelona Africaines (CEDA) Sales & Circulation Department Loita 0000Baglr Tw oseld en,:9 04 8 P. 541 302 Seabird House Mezzanine 1 5 garae Town 08009 Conselom Cn Abidjan 04 22-28 Wyndham Street, Control PDO. Boo 68077 haa Fa 33 Tel: (i4sa) 4349 AbidJun 04 ~~~~~~~Hong Konig, China Nairobi P o 33 80 acln Tel: (225) 24 6510246511 Tel: (852(230-10 Tel (420853. 221426 Karachi-75350 Fax: (343) 488-3421 Fax: (225) 25 0567 ~ Fa: (361) 230510903 Tel: (8254 2) 536-SS TE-al: (92 21) 446307 E-mail. baL@rcelonakudirna. es F ( 5Fan: (852) 2526-1107 Fax: (254) 2-330854, 561654 Tel: (92 21) 45476307 CYPRUS E-mail: salese'asia2o00.com.hk E-mail: Legacy@form-net.com Fan: (92p21) 4547640 E-maI bArcA,onaE'MAnDIpES Center for Applied Research URL: http:llwwwasia2000 com.hk KOREA, REPUBLIC OF E-al upkTefient SLaNkA, THu e MALksh Ep Cyprus CoIIeO HUNGARY Dayang Books Trading Co. Pak Book Corporation 1 00, Sir Chittampalam Gardiner 6. Diogenes Street. Engomi Euro Info Service International Division Aziz Chambers 21, Queen's Road Mawatha Niosiax 00 Margtzgeti Europa Haz 783-20, Pangba Bon-Dong, Lahore Colombo 2 Tel:si (37)903 H-i 538uaps Socsa-ku Tel: (92 42) 636 3222: 636 0885 Tel: (94 1(32105 Fan: (357 2) 66-2051 Tel: (36 1(350 80 24, 350 80 25 Seoul Fax: (92 42) 636 2328 Faa: (94 1) 432104 Fan: (36 1) 350 90 32 Tel: (82 2) 536-9555 E-Mail: pbc@Dbrain.net.pk E-mail. LHLS'sri.lanka.net E-mail: euroinfora'mail.matav hu Fan: (82 2) 536-0025 E-mail: searnapffchollian.net Recent World Bank Technical Papers (continued) No. 449 Keith Oblitas and J. Raymond Peter in association with Gautam Pingle, Halla M. Qaddumi, and Jayantha Perera, Transferring Irrigation Management to Farmers in Andhra Pradesh, India No. 450 Andres Rigo Sureda and Waleed Haider Malik, eds., Juidicial Challenges in the New Millennium: Proceedings of the Second Sulmmit of the Ibero-American Sutpreme Couirts No. 451 World Bank, Privatization of the Power and Natural Gas Industries in Hungary and Kazakhstan No. 452 Lev Freinkman, Daniel Treisman, and Stephen Titov, Suibnational Budgeting in Ruissia: IPreempting a Potential Crisis No. 453 Bartlomiej Kaminski and Michelle Riboud, Foreign Investment and Restructuiring: The Evidence from Huingary No. 454 Gordon Hughes and Julia Bucknall, Poland: Complying with EU Environmental Legislature No. 455 Dale F. Gray, Assessment of Corporate Sector Valute and Vuilnerability: Links to Exchange Rate and Financial Crises No. 456 Salman M.A. Salman, ed., Grouindwater: Legal and Policy Perspectives: Proceedings of a World Bank Seminar No. 457 Mary Canning, Peter Moock, and Timothy Heleniak, Reforming Education in the Regions of Ruissia No. 458 John Gray, Kazakhstan: A Review of Farm Restructuring No. 459 Zvi Lerman and Csaba Csaki, Ukraine: Review of Farm Restrulctllring Experiences No. 460 Gloria La Cava and Rafaella Y. Nanetti, Albania: Filling the Vullnerability Gap No. 461 Ayse Kudat, Stan Peabody, and Caglar Keyder, eds., Social Assessment and Agricultutral Reform in Central Asia and Tuirkey No. 462 T. Rand, J. Haukohl, and U. Marxen, Municipal Solid Waste Incineration: Requiirementsfor a Successfiul Project No. 463 Stephen Foster, John Chilton, Marcus Moench, Franklin Cardy, and Manuel Schiffler, Groundwater in Rutral Development: Facing the Challenges of Supply and Resource Sustainability No. 465 Csaba Csaki and Zvi Lerman, eds., Struictural Change in the Farming Sectors in Central and Eastern Europe: Lessonsfor EU Accession-Second World Bank/FAO Workshop, June 27-29, 1999 No. 466 Barbara Nunberg, Readyfor Europe: Putblic Administration Reform and Euiropean Union Accession in Central and Eastern Europe No. 467 Quentin T. Wodon with contributions from Robert Ayres, Matias Barenstein, Norman Hicks, Kihoon Lee, William Maloney, Pia Peeters, Corinne Siaens, and Shlomo Yitzhaki, Poverty and Policy in Latin America and the Caribbean No. 469 Laurian Unnevehr and Nancy Hirschhom, Food Safety Issues in the Developing World No. 470 Alberto Vald6s, ed., Agricuiltutral Support Policies in Transition Economies No. 471 Brian Pinto, Vladimir Drebentsov, and Alexander Morozov, Dismantling Ruissia's Nonpayments System: Creating Conditionsfor Growth No. 472 Jit B. S. Gill, A Diagnostic Frameworkfor Revenute Administration No. 473 Esen Ulgenerk and Leila Zlaoui, From Transition to Accession: Developing Stable and Competitive Financial Markets in Builgaria No. 474 loannis N. Kessides, ed., Huingary: A Regutlatory and Structural Review of Selected Infrastructure Sectors No. 475 Csaba Csaki, Zvi Lerman, and Sergey Sotnikov, Farm Sector Restructutring in Belarus: Progress and Constraints No. 476 Katherine Terrell, Czech Republic: Labor Market Report No. 481 Csaba Csaki, John Nash, Achim Fock, and Holger Kray, Food and Agricultuhre in Builgaria: The Challenge of Preparing for EU Accession No. 482 Peter Havlik, Trade and Cost Competitiveness in the Czech Republic, Huingary, Poland, and Slovenia No. 483 Mojmir Mrak, Commutnal Infrastructutre in Slovenia: Suirvey of Investment Needs and Policies Aimed at Encouiraging Private Sector Participation No. 484 Csaba Csaki and Laura Tuck, Rural Development Strategy: Eastern Euirope and Central Asia No. 488 Nina Bubnova, Governance Impact on Private Investment No. 489 Tim Schwarz and David Satola, Telecommuinications Legislation in Transitional and Developing Economies No. 490 Jesko Hentschel and Radha Seshagiri, The City Poverty Assessment: A Primer No. 492 Tuntivate Voravate, Douglas F. Barnes, and V. Susan Bogach, Assessing Marketsfor Renewable Energy in Rutral Areas of Northwestern China THE WORLD BANK 1818 H Street, N.W. NVashington, D.C. 20433 USA Telephone: 202-477-1234 Facsimile: 202-477-6391 Internet: ww-v.worldbank.org E-mail: feedback@worldbank.org ISBN 0-8213-4963-5