Report No. 25923-UZ Uzbekistan Living Standards Assessment Policies To Improve Living Standards (In Two Volumes) Volume I: Summary Report May 2003 Human Development Sector Unit Europe and Central Asia Region Document of the World Bank CURRENCYAND EQUIVALENTUNITS (ExchangeRates effectiveApril 1,2003) Currency Unit = Soum US$l = Soum 968.42 (officialexchange rate) US$l = Soum 969.64 (OTC exchange rate) US$l = Soum I,000.00-1,020.00 (consumer goods/exchange bureauexchangerate) US$l = Soum 1,120.00-1,140.00 (parallelmarket exchangerate) ACRONYMS AND ABBREVIATIONS AIDS AcquiredImmune Deficiency Syndrome GDP GrossDomesticProduct BEEPS BusinessEnvironmentand Enterprise ILO InternationalLabor Organization Performance Survey CEE Centraland Eastern European IMF InternationalMonetaryFund CIS Commonwealth of IndependentStates LFS LaborForce Survey CTS CommonTariff Schedule MAWR Ministry of Agricultureand Water Resources DHS Demographicand Health Survey NGO Non-GovernmentalOrganization EBRD European Bank for Reconstructionand PPP PurchasingPowerParity Development EU EuropeanUnion PTA Production and TradeAssociation FSU Former SovietUnion SME Small and Medium-Size Enterprises FBS Family Budget Survey TB Tuberculosis 1 Vice President Johannes Linn (ECAVP) 1 Country Director Dennis de Tray (ECCOS) SectorDirector Annette Dixon (ECSHD) Acting SectorManager Mansoora Rashid (ECSHD) Task Team Leader Reema Nayar (ECSHD) 2 ACKNOWLEDGEMENTS This report is the outcome of substantial collaboration between a Government o f Uzbekistan team and World Bank team. We are extremely grateful to all our counterparts in the Government of Uzbekistanfor their keen interest and valuable inputs throughout the process of preparation of this report. The comments and inputs provided by the Government Working Group to both volumes of the report, have been reflected to the extent possible. The final views expressedinbothreports, however, arethose of the World Bank team. The Government Working Group consists o f a Policy Level Group that was chaired by Mr. Furkat Baratov (Deputy Minister of Economy), and includes Mr. Rustam Akhliddinov (First Deputy Minister o f Public Education), h4r. Botir Alimukhamedov (Deputy Minister of Labor and Social Protection), Mr. Durbek Akhmedov (Director, Center of Effective Economic Policy), Mr. Damin Asadov (First Deputy Minister of Health), Mr. Abduvohid Juraev (Deputy Minister of Agriculture and Water Resources), Ms. Reyhanat Makhmudova (Deputy Head of the State Statistics Committee), Mr.Zavikhulla Nasretdinkhodjaev, (Head of Division, Information-Analytical Department, Cabinet o f Ministers), Mr. Khurram Teshabaev (Director, Center o f FiscalPolicy), Mr.Shukrat Vafaev (Head of Division, Cabinet of Ministers). The Technical Level Group was chaired by Mr. Bekzod Musaev (Ministry of Finance), and includes Mr. Tursun Akhmedov (Center of Effective Economic Policy), Ms. Lyudmila Babadjanova (State Statistics Committee), Ms.Karimova (Center o f Effective Economic Policy), Ms. Zulfia Khamidova (Center o f Fiscal Policy), Mr.Bakhtiyor Khoshimov (Ministry o f Health), Mr. Valeriy Lapyngin (Minstry of Economy), Ms. Zulkhumor Mutalova (Ministry o f Health), Mr. Alexander Novotny (Ministry o f Labor and Social Protection), Mi.Artur Tukhtarov (Information and Analytical Department, Cabinet of Ministers), Mr. Erkin Yusupov (Ministryof Agriculture and Water Resources), Mr.Anvar Zokirov (Ministryo f Public Education), Mr.UlugbekZakirov (State Statistics Committee). The World Bank team was led by ReemaNayar. Both volumes ofthe report are based on backgroundpapers prepared by Dilnara Isamiddinova, Brian Kropp, John C. Langenbrunner, Mark Lundell, Caralee McLiesh, Kalpana Mehra, Michael Mertaugh, Edmundo Murrugarra, Yoshimi Nishino, Anahit Poghosyan, Mansoora Rashid, Kinnon Scott, Bekzod Shamsiev, Andriy Storozhuk, William Martin Tracy. Additional contributions were received from Bekzhod Abdurazzakov, Mariam Claeson, Dina N. Abu-Ghaida, Arvo Kuddo, Manju Rani and Moukim Temorov. Peer Reviewers for the report were Peter Lanjouw and Ana Revenga. The team i s also grateful to Mansoora Rashid, Maureen Lewis, Annette Dixon, Dennis D e Tray, David Pearce for their support and advise, and to Daniella Gressani, Robert Anderson, Jan Rutkowski, Pradeep Mitra, Chris Lovelace and Ritu Anand for helpfulcomments and discussions. Helpful discussions are also acknowledged from various members of the donor community (ADB, UNDP, USAID), local government authorities, and non-governmental research organizations. We thank the Multi-Sector Team Learning Program at the World Bank for providing financial and other support to further the collaboration between the World Bank and Government of Uzbekistan Teams. We would also like to acknowledge supplementary financial support for parts o f the research from the World DevelopmentReport (2002), the Health Sector Decentralization Thematic Group, and a Japanese Trust Fund. 3 UZBEKISTAN LIVINGSTANDARDSASSESSMENT: POLICIES TO IMPROVE LIVINGSTANDARDS I.INTRODUCTION 1. Poverty was not unknown in Uzbekistan prior to the transition. At independence, the poverty rate in Uzbekistan was much higher than most o f the other Former Soviet Republics: a reflection of Uzbekistan's low gross per capita income and relatively high levels of inequality. However, as a result of its socialist legacy, Uzbekistan's education and health achievements compared favorably to countries with similar levels of income. 2. Since independence, Uzbekistan has followed a distinct economic strategy, entailing gradual transformation of the economy, while emphasizing social stability. The "Uzbek Model" o f development has focused on developing industrial and manufacturing capacity in a predominantly agricultural economy, using direct and substantial state guidance. An important objective o f the strategy has been to raise living standards and expand employment opportunities, while protectingvulnerable groups against abject poverty. 3. Has this approach alleviated the problems o f poverty inherited by the country? This report provides the first national levelpicture of living standards inUzbekistan since independence. It has three main goals: (i)to examine the current status of living standards in Uzbekistan (ii)to identify key challenges and constraints to improvingliving standards, and (iii) to suggest priority policy actions that are neededfor broad based improvements in living standards inthe country. The report was prepared in close collaboration with a working group from the Government o f Uzbekistan, to ensure the relevance o f the findings and to build capacity for analytical evaluation of living standards usinghousehold survey data. 4. The report is based on the recently improved and nationally representative Family Budget Survey (FBS) (2000/01) carried out by the Uzbek statistical authorities. Since 2000/01 was the first year of implementation o f the revisednationally representative survey, the new survey i s considered by the Statistical Authorities to be a pilot. This is also the first time the data have been used for poverty analysis, and the exercise has yielded important feedback for further strengthening the survey. Despite these important caveats, the FBS does provide the first comprehensive information on living standards inthe country, and represents the best available information at this time. Results that appear to contradict conventional wisdom cannot be rejected a priori, since they represent the responses o f about 10,000 households. They must be verified with future rounds of the survey as well as special studies. Inaddition to the FBS, the report uses other sources o f information, including surveys o f firms, farms, institutions and individuals, as well as administrative data. The study also uses international evidence to compare and contrast Uzbekistan's living standards and policy outcomes relative to other countries, including CEE (Central and Eastern European) countries and other CIS (Commonwealth of Independent States) countries. 5. A broad measure of household welfare encompassing both income and non-income dimensions of poverty is used in this report. To determine income poverty, the report uses an income threshold, or poverty line, that reflects minimum subsistence (see Box 1). Households with income below this threshold are considered poor; and those with income above it, non-poor. The report also uses two key poverty measures: the first measure, the poverty rate (or incidence of poverty), is the proportion o f a particular group that is income poor, and represents the risk o f poverty faced by the group. However, a group may have a low poverty rate, but becauseof its large population share may comprise a large share o f the total poor population. Therefore, the proportion of poor in a particular group relative to the total poor population is the second main indicator used in this report. The non-income well-being indicators used in the report include capabilities, such as health and education, basic services (sanitation, water), the vulnerability to risks and shocks, as well as voice and participation inthe society. 4 6. This report comprises two volumes. This first volume provides a summary o f the findings and key policy recommendations of the report, preceded first by a brief overview of the key messages. The second volume contains the more detailed technical analysis on which this first volume i s based. Box 1: Measuring Poverty in Uzbekistan To measure material well-being, typically a measure of welfare based on total consumptionis used. The data from the Family Budget Survey only permitted the constructionof a reliable food consumption aggregate, not a total consumption aggregate. This food consumptionaggregate (adjusted for regional and monthly differences in prices) was compared with two food poverty linesto provide estimates of total poverty and extremepoverty. The analysisdrew on the findings of Olson Lanjouw and Lanjouw (2001) which establish that under certain theoretical assumptionsthe poverty rate as measuredby a head count measure would be the same under the case where a measure of food poverty is compared to a food poverty line and one where atotal consumptionmeasure is comparedto atotal poverty line. This approachwas found to yield reasonable results both for the headcount index and for the profile of poverty in several practical cases in Latin America and Asia. Its reliable empirical performancewas further established for Kazakhstanand the Kyrgyz Republic (Scott and Mehra, 2002). The approach permits the best possible use of FBS data to understand the extent of poverty in Uzbekistan and the characteristics of the poor. It does not, however, permit estimates of inequality, depth and severityof poverty. Moreover, the profile of povertymay yield less pronounceddifferencesbetweengroupsthan would otherwise bethe case. The food poverty line was establishedby estimatingthe cost of a consumptionbasket that would provide 2100 calories per personper day. The consumptionbasket is basedon actualconsumptionpatterns of thepoorpopulation and is convertedinto soums usingthe pricesfaced by the poorerpopulation. A secondpoverty line, to characterizeextreme poverty, was computed based on a caloric intake of 1500 calories per person per day. The value of the food poverty line thus obtained, in October 2000 prices, is 3601 soums per moth, which is 50% higher than the value of the minimum wage in 2000, while the extreme poverty line (2572 soums) is quite close (just 5% higher) than the minimum wage inOctober 2000. It is importantto notethat all poverty estimationwas donejointly with the Government's inter-ministerial working group for the Living Standards Assessment. In addition the sensitivity of the profile to assumptions on the differences in needs of different householdmemberswas testedusingweights developedby the Ministry ofLabor and SocialProtection. References: Olson-Lanjouw and Lanjouw, 2001, "How to compare apples and oranges" Poverty Measurement Based on Different DefinitionsofConsumption", Reviewof IncomeandWealth, Series 47, Number 1,March2001. Scott andMehra,"A Descriptionof Poverty in Uzbekistan," Backgroundpaperpreparedfor LivingStandards Assessment BriefOverview of Main Findings 7. There have been improvements in some indicators of well-being as a result of recent economic growth; but poverty remains a serious challengefor Uzbekistan. According to administrative data, some indicators o f well-being (for example, life expectancy, infant and child mortality rates, enrollment rates in basic education) have recovered or improved substantially with the resumption o f economic growth; illiteracy has been eliminated; many households now own their dwelling; and rural households have access to household plots. However, income poverty remains a problem, over a quarter of thepopulation (27.5%) can be characterized as poor and a third of them extremely poor. As discussed below, some non-income dimensions o f well being are also of concern: survey data show much higher infant mortality rates compared to administrative sources', infectious diseases such as tuberculosis and AIDS are increasing, and some educational achievements are at risk. 1. Income Poverty 8. Poverty is predominantly rural, with higher concentration in particular regions. Like other CIS countries, a large share-nearly 70 percent of all poor-reside inrural areas. Among regions, Kashkadarya, ' 75%higher in survey relativeto administrativedata in 1996. This is not uncommonin FSUrepublics.Discrepancies are even higher in some other republicssuch as AzerbaijanandTajikistan. 5 Namangan and Karakalpakstan have poverty rates that are much higher than the national average, and their inhabitants comprise almost 40 percent of all poor. 9. Employment has not protected individuals against poverty. Nearly half o f all households with employed household heads are poor he rural poor are heavily involved in agriculture, while the urban poor work disproportionately in industry, where state ownership is high. The poor are also more likely to work in the informal sector. Very low and irregular wages and high wage inequality (much higher than CEE Countries, and similar to some CIS countries) i s the main reason for poverty among workers. The problem is most severe in agriculture and explains the high incidence o f rural p o v e d . Low and irregular wages are the result o f continued state subsidization and lack o f restructuring o f insolvent enterprises. Low agricultural wages also reflect implicit taxation of agriculture, and heavy state involvement in agricultural product and input markets. The private sector, essential for employment and wage growth, is heavily constrained by a poor business climate, inflation, large administrative and regulatory burden, limited financing opportunities and informal payments. As in other CIS countries, excessive wage inequality is explained only to a small extent by productivity differentials among workers, and reflects wage arrears, regional differences, and informal activity. Low and uncertain remunerationfor employment is one reason that poverty rates among the employedare very similar to households with inactive heads. 10. Basic and secondaly education do notprotect againstpoverty; only higher educationpays. Most of the poor have heads who have completed basic or secondary education. Consistent with evidence from other CIS countries, the Family Budget Survey shows that only higher education i s correlated with low poverty rates, while households headed by secondary and basic education graduates have rates o f poverty that are no different than the national average. This reflects an early transition labor market where earnings differences do not fully reflect productivity differences. Only higher education graduates command a significant wage premium in the labor market. In contrast, in CEE countries, secondary education does command a wage premium and is therefore associated with lower poverty. Insufficient demand for the skills of secondary graduates could additionally explain the lack o f a wage premium. 11. Many poor are inactive, reflecting insufficient employment opportunities and incentives to retire early. The inactive are mainly pensioners and working age women. Inactivity among women is very high, comparable to Turkmenistan and Azerbaijan, and i s mainly attributable to a lack o f higher education, more young children and cultural factors. According to official data, the employment rate (employed/working age population) has declined steadily over the last decade, despite the recent growth. Job opportunities for the growing working age population have been limited. Most inactive poor have coped with poverty through informal sector activity, access to plots, and social benefits. 12. The unemployedare aparticularly vulnerablegroup. The unemployment rate inUzbekistan is about 6 percent3,with a highrate o f unemployment among youth. Though they comprise a small share o f the poor, the unemployedhave the highest poverty rate amongthe working age population. All sources indicate that agricultural wages are low. The Family Budget Survey reports that the incidence of wage arrears is highest in agriculture, followed by construction and industry. Gini coefficients for wages according to the Family Budget Surveywere over 0.5; intemational comparisons are based on latest available data 'According to the FBS. According to official statistics, registeredunemployment at the end of 2001 was 0.4% of economically active population, and survey based unemployment from the new labor force monitoring surveys carried out by the Ministry of Labor and Social protection is nearly 4%. 6 2. Non-IncomePoverty 13. Health outcomes show vulnerability,for the poor, but also for the non-poor. There have been recent improvements in the incidence o f infectious diseases; however, the incidence o f some preventable infectious diseases (tuberculosis and AIDS) have increased. Death rates from chronic diseases increased until the mid-l990's, and remain high4.Poor households, particularly in rural areas, face a higher incidence of infectious disease and worse nutritional outcomes. The reduction in public spending on health, its disproportionate targeting to the non-poor (through high spending on hospitals), increasing formal and informal costs of health care, the uneven quality of the health care system, and low household income are possible explanations. Despite reportedly improved coverage o f rural populations with running water, the lower access to basic sanitation and safe water also contributes to the low health status of rural populations. 14. Some educational achievements have eroded, although recent trends are encouraging. The net enrollment ratios in basic education increased after the mid-1990's to about 95% in 2000/01. However, coverage rates at other levels - e v e n in higher education (in contrast to other transition countries), remain below pre-transition levels. In very recent years these negative trends have started to reverse. Educational outcomes, particularly at these levels are worse for the poor households and in particular regions. As in the case o f health, the reduction in public spending for education, its disproportionate targeting to the non-poor (particularly at higher levels o f education), increasing formal and informal costs of education; uneven quality and lower perceived relevance of education, and low household incomes (and higher opportunity costs of education) are possible explanations. 15. Socialprotectionprograms show mixed success inprotecting against risks. The Government has an extensive social protection system, to protect the most vulnerable groups. The pension system, the largest (in terms o f expenditure and coverage) o f all social protection programs, has helped alleviate poverty among pensioners. However, this protection has come at large current costs, e.g. highpayrolltaxes, and incentivesto early retirement, and the system faces large future fiscal costs due to population aging. The targeting o f cash assistance programs by the Mahallahs shows mixedresults and can be strengthened, while the privileges and subsidies largely benefit the non-poor. Unemployment benefit programs have very little impact (given low coverage) and mostexpenditures are skewed towards cost-ineffective active labor market programs. 3. Policiesto ImproveLiving Standards 16. Policies to improve livings standards inUzbekistanwill need to focus on five key areas. A. Sustained economic growth will require facilitation of private sector growth and, given large rural poverty, agricultural sector reforms: Facilitation of private sector and SME development. Key actions include macroeconomic stability and proceeding with fundamental structural reforms (such as liberalization o f the foreign exchange and trade regimes, financial sector reforms), imposition o f hard budget constraints on state enterprises, reducingthe burden of tax and regulations. Restructuring of the state sector and easing o f barriers to private sector development is critical for improving the allocation o f labor and capital, and promoting growth in productivity and employment. Improvements in agricultural productivity and incomes. Key actions include improving the incentive structure in agriculture through price liberalization and reduced state intervention in product and input markets, and increasingequity o f land reforms. The recent experience of the Kyrgyz Republic, shows that The Government reports n o further increases in the incidence o f chronic diseases in the past five years. 7 major agricultural reforms in the area of land policy, the incentive framework, credit, irrigation and agricultural services, lead to significant agricultural growth and reduction inrural poverty. 0 Distributional impact of reforms. The above set of reforms will significantly improve the living standards o f the country as a whole (including the poor) though some groups may be adversely impacted inthe short run.These individuals will needto be protected by a well targeted safety net(see below). B. Improving capabilities in health and education will be important for improving welfare of the population and will also allow households to benefit from economic growth. Policies to improve capabilities include: 0 Improving education outcomes, will require the provision o f relevant, transferable and general skills to the population and targeted actions to improve access among the poor and in poor regions. Implementation of these policies will require increased resource allocation to poorer regions ,as well as continuation of ongoing efforts to provide textbooks and educational materials to needy families. It will also be important to increasewomen's enrollments inhigher levels of education. 0 Improving heaZth status will require continued actions to reduce infectious diseases and improve nutritional outcomes, with a particular focus on prevention, by increasing relative allocations to primary health care (particularly in poorer regions) through acceleration of hospital restructuring efforts; strengthening the monitoring and targeting o f existing exemptions to the poor for care at all levels (including actions to reduce informal payments); and continued attentionto improvingaccess o f the rural poor to water and sanitation services. It will also be important to introduce mechanisms to protecting the non-poor from impoverishment due to catastrophic or chronic illness. C. Providing effective social protection will require reducing the employment and future fiscal costs associated with the social insurance system and improving incentives to work (through pension reform), strengthening unemployment programs, and further improving the targeting of cash social assistance and child allowance programs, and phasing out privileges. To reduce social costs of restructuring and other reforms, labor redeployment programs could be usedto assist laid off workers, and existingsocial assistance delivery models could be strengthenedand usedto target subsidies to individuals who may face cost of living increases from economic reforms. D. Building inclusive institutions and reducing informal payments (which constitute a disproportionate burden for the poor), will allow a larger share of the populationto benefit from economic and social policies. Actions include increasing accountability of state institutions (through improved quality o f regulations, introduction o f performance based salary structures, and strengthened oversight and monitoring), increasing transparency and access to information, strengthening the voice o f communities, and encouragement o f civil society organizations. E. Improving poverty monitoring will require continued implementation ofhousehold based surveys such as an improved Family Budget Survey, to track changes in household welfare, the implementation of regular Labor Force Surveys to measure labor market developments, the updating of the population census. It will be important to provide data access to the public and to strengthen the capacity of poverty analysis within and outside the government. A well organized and regular system for gathering, disseminating and evaluatingdata will helpcreate sound policies and programsfor improvingliving standards inthe country. We now turnto a more detailed discussion o f these results. 8 11. LIVING STANDARDS INUZBEKISTAN 1. GrowthandLivingStandards 17. Uzbekistanrealized the "mildest" transition recession which may haveprevented large real income declines in the early transitionyears. Although comparable data are not available over time to assess income poverty, the smaller fall in incomes was likely associated with smaller declines in living standards compared to many other CIS countries. Recently, Uzbekistan has seen six consecutive years of positive GDP growth, averaging 4% per year according to official figures.5 18. Some indicators of well-being show improvements associated with Uzbekistan's recent growth. According to administrative data, some indicators o f well-being (for example, life expectancy, infant and child mortality rates, participationrates inbasic education) have recovered or improved substantially with the resumption o f economic growth. Other gains that had started inearlier decades, seem to have continued since independence: illiteracy has now been completely eliminated, and gender gaps in educational achievement have been closing and remain primarily at higher levels. Reflectingthe early privatization o f housing, 95% of all households own their dwellings, and the majority o f rural households have access to household plots for cultivation. 19. However, serious challenges remain. Growth has not generated sufficient employment opportunities or substantially improved incomes o f the population: 20. Thegrowth rate in Uzbekistanlags behind most other CIS countries. Since the end o f the transition recession inCIS countries, Uzbekistan's average GDP growth has been the lowest, and cumulative growth in per capita GDP has beenthe second lowest. Growth has not beenbroad based; accordingto official statistics, duringthe period from 1996-2002, real value added inagriculture grew on average at about 3.3% per annum, and in industry it grew at only 2.0% per annum. Growth was higher in the trade sector, which however, accounts for only a small share o f employment. Thus, incomes and wages remain low. In 2001, GDP per capita at purchasing power parity was $2,440, third lowest among CIS countries, and the average monthly wage was $29 at the official exchange rate (or $17 at the parallel market rate)6. 21. The growth in productivity has been limited. Labor productivity7 in industry has grown at only 0.8% per annum in 1996-2002, in agriculture at about 5.5% per annum, and services at 1.6% per annum. The limited productivity growth reflects insufficient restructuring (particularly outside of agriculture) and associated movement o f resources to high productivity activities in the private sector and in services and trade. Although some restructuring has been initiated in the agricultural sector, Uzbekistan has not seen the dramatic sectoral shifts in output or employment witnessed in Central and Eastern European countries, and public services still dominate the services sector. Private sector share o f GDP is among the lowest in transition countries and over 60% of investments come from the state sector. Foreign direct investment flows have been falling steadily, and at an average o f $3 per capita in 2000-2002 are among the lowest in CIS countries. Box 2 shows the importance o f productivity growth for increases in incomes and wages o f workers. Alternate IMF estimates of growth have been lower, at abouttwo thirds of this rate. Dynamics of US dollar-denominated income indicators varied between 1996 and 200112002. GDP per capita inPPP terms steadily increased (from $2,000 in 1996)while US$GDP per capita at current exchange rates declined (from $601 in 1996 to $384 in 2002 at the official exchange rate, and from $409 to $226 at the black market rate). Average wage during the same perioddeclined from $54 at the official exchange rate and from $37 at the black market rate 'Labor productivity figures are World Bank staff calculations basedon official national accounts and employment statistics. 9 Box 2: Economic Growth, Productivity and Living Standards during Economic Take-off The experience of countries that have successfully industrialized while achieving substantial improvement in living standards demonstratesthe importanceof intensive, productivity-ledeconomic growth. For example, as the table below shows, Spain and Korea whose income levels and agriculture-dominatedeconomies in mid-1960s were not dissimilar to those of Uzbekistantoday, achieved and sustained remarkableGDP growth rates. They achievedthese not only by encouraging high levels of savings and investment, but also by openingup their economiesandpursuingexport-orienteddevelopment strategies which increasedcompetitionand efficiency as well as stimulated high productivity growth. Productivity growth was particularly strong in industry, the sector most open to foreign competition. This enabled the industrial sector both to absorb labor released from agriculture as its productivity grew, and to ensure steady improvementinliving standards throughout the economy as the benefitsof better manufacturingwages was transmittedto other Average annual growth rates Spain Korea Korea Memo: Uzbekistan (unless indicatedotherwise) 1964-74 1965-80 1980-89 1996-2002 Total GDP: 6.4 9.9 9.7 4.0 Agriculture 2.5 3.0 3.3 3.3 Industry 9.1 18.7 13.1 2.0 Services 5.4 9.6 9.1 4.5 Value added per worker Agriculture 8.4 0.5 19.0 5.5 Industry 8.7 16.6 16.3 0.8 Services 2.7 11.8 12.5 1.6 sectors. As a result, during the 10-15 year take-off period inthe two countries, employment generationexceeded populationgrowth rates andper capitaincomes (as measuredby Atlas GNIper capita) rose by 4-14 times. In contrast, Uzbekistan's more inward-oriented development approach, while ensuring steady if modest overall economic growth during 1996-2001, has been less successful in creatingenvironmentfor productivity growth. Although investmentin Uzbekistanhas been high (averaging 30 percent of GDP during 1996-2001), its efficiency - payoff in terms of higher productivity and economic growth- has been relatively low, as the domestic producers' incentives to improve have been blunted behind the wall of protection through foreign exchange and trade restrictions. Productivity performancein industry, a major recipient of investmentand subsidies, compares particularly unfavorably with other industrializingcountries during their "take-off' periods. Productivity growth in industry has also been considerably lower than that in agriculture, which itself has been below potential, in part due to an implicit resource redistributionby the state away from agriculture during most of the period. This lack-luster productivity performance is an important reason why Uzbekistanhas not seen a marked improvementin living standards despite economic growth since mid-l990s, and why employment growth still lags behindpopulationgrowth. A decisivepolicy shift towards a more open economy, removalof barriersto foreign trade, competition and private sector development as well as macroeconomic stability would be required for Uzbekistanto realize its economic potential and achieve income gains similar to those of Spain, Korea and other countries that successfully industrialized. Source: World Bank, Uzbekistannational statistics authorities. 22. Growth has not generated sufficient employment opportunities for the growing population. Although employment inabsolute terms has increased according to official statistics, recent economic growth has not translated into a sufficient number of productive jobs for the growing working age population. The employment rate (ratio o f employment to working age population), based on official statistics, has declined steadily since independence (by 12 percentage points between 1991 and 2002), and recent growth has not reversed this trend. Activity rates are particularly low among women (only 33%, comparable to Turkmenistan, Tajikistan, and Azerbaijan). Open unemployment (6%), high youth unemployment (13%), underemployment, the significance o f temporary work arrangements, and informal sector employment are all additional manifestations o f inadequate utilization o f labor resources8. The experience o f East Asian countries, where a growing workforce was accompanied by faster GDP growth per capita, demonstrates the importance of tapping the potential of a growing workforce. 23. Income inequality hns increased and has prevented n higher impact of growth on poverty reduction. As in the case o f poverty, comparable data on income inequality are not available over time for Uzbekistan. However, wage inequality has increased, particularly between workers in the agricultural sector Numbersquotedhere are from Family Budget Survey which differ quantitatively,althoughnot qualitatively, from official labor force monitoring. 10 and all other workers, suggesting that rural-urban disparities have widened'. While the average agricultural wage was comparable to that in industry in 1991, by 2002 it was only 21%. Unlike the initial increases in inequality in Central European transition countries, this increase in inequality has not been a result of alignment o f wages with productivity, as agricultural productivity growth has been higher than that in industry. Levels of wage and income inequality in Uzbekistan today are comparable or higher than those observed in other CIS countries, and significantly higher than those in countries o f Central and Eastern Europe andthe Baltics. Figure I: Thepoor arepredominantly rural, andpoverty rates differ significantly across regions. GeographicDistributionof Poverty 80 - Share of population -Share of poor -Incidence of poverty- - - - --Poverty Incidence -National1 2. A Profile of Living Standards 24. Poverty remains a significant problem and an estimated 27.5% of the population, or 6.8 million people in Uzbekistan are unable to meet basic consumption needs. Approximately one-third of all poor households can be classified as extremely poor. As discussed below, income poverty is associated with other forms o f deprivation, and some non-income dimensions o f welfare are under stress. A. Rural and RegionalPoverty 25. Poverty is predominantly a rural phenomenon. Approximately 4.5 million poor people, or 70% o f Uzbekistan's poor live in rural areas. Rural populations are 35% more likely to be poor and 58% more likely to be extremely poor than their urban counterparts. 26. There are large regionalpockets ofpoverty: Kashkadarya and Namangan have the highest rates of poverty and extremepoverty. Individuals living inKashkadarya are seven times more likely to be poor than those living in Tashkent city. Kashkadarya itself accounts for a fifth of Uzbekistan's poor and nearly 40% o f its extreme poor (Figure 1). While education, health and employment account for some o f these regional differences, they do not account for all of them. Other factors, such as differences in infrastructure or other endowments, policies or institutions are likely to be important. Workers in agricultural cooperativesadditionally receive a share of profits- when these are made- as well as in-kind payments; to that extent wage figures alone may overstate disparities. Figure 2: Rural Populationshave limited access to many basic services Access to Public Service and Utilities by Geographic Location. RunningWater Central Hot Central Heat Central Sewage Natural Gas Telephone Water EIUrban .Rural ' 27. Rural and some regional populations have lower access to basic services and amenities such as systems for runningwater, central heat, hot water, and sewerage. Official data show an increase in access to water supply for rural populations and is, at present 78% for the republic. The Family Budget Survey, however shows that rural populations have low access to many basic services. For example, only a quarter of the rural population has access to runningwater and less than 2% to central sewerage (Figure 2). Moreover, households in particular regions (such as Karakalpakastan, Khorezm) have been affected by limited water availability and drought; the poor are likely the hardesthit". The capabilities of the population, such as health and education status, are worse inrural areas, and inparticular regions B. Labor MarketsandPoverty The Working Poor 28. The employed have a lower incidence of poverty than the inactive or unemployed, but this dqference is not very large (Figure 3). In contrast, inthe more advanced transition countries in Central and Eastern Europe and the Baltics, the employed have a much lower rate o f poverty than the inactive; and poverty i s much more closely related to unemployment and inactivity. About half o f the poor are employed. A large share of the ruralpoor (more than half) are employed inagriculture; this share is substantially higher when employment in agricultural industries and processing is included. Private, non-farm employment accounts for less than 10% o f rural non-agricultural employment. The urbanpoor work disproportionately in industrywhere state ownership is high. loGovernment data show a steady increase inwater supply provisionto some of these ecologically disadvantaged areas relative to the early 1990's. 12 Figure 3: Thepoor are equally divided between the inactive and the working Poverty and EmploymentStatus Inactive Employed Unemployed Share o f Population 0Share ofPoor +Incidence o fPoverty i 29. Wages are low and uncertain, and wage inequality is high. Rural agricultural workers have the lowest wages among all sectors and have a very high incidence of wage arrears. The gini coefficient of median hourly wages i s 0.59 in the private sector and 0.52 in the public sector. High wage inequality in Uzbekistan likely reflects the significance of informal employment and wage arrears. There is a strong association between poverty and the incidence o f wage arrears, seasonal variation in employment opportunities, and casual or temporary work arrangements. Thus thepoorface greater vulnerability in labor market outcomes. 30. Participation in the informal sector is linked topoverty. Estimates of the size o f the informal sector (20% to 50% depending on the definition) are similar to Kazakhstanand higher than in the Kyrgyz Republic, but lower than in the Caucasus (Armenia, Azerbaijan, Georgia). Although the informal labor market i s an important source o f employment, it is typically associated with greater insecurity in labor market status, reduced trans arency, and with exclusion from government social programs. It is also associated with lower productivity IPand with increased difficulties in tax collection and therefore fewer resources available for redistribution or basic social services. The UnemployedandInactive 3 1. The unemployed have the highest incidence of poverty, but constitute a small share of all poor. The unemployment rate is generally higher in poorer regions, although the correlation is not perfect. Unemployment was highest in Samarkhand, Dzizhak and Surkhandarya. Combined with substantially higher population pressure in rural areas, limited availability o f employment opportunities has manifested itself in higher rural unemployment, greateryouth unemployment, and greater seasonality of employment. 32. The inactivepoor are mainly pensioners. Half o f the poor live in households whose heads are not active in the labor market. The majority of the "inactive" group are pensioners: households headed by pensioners account for just under a third o f the inactive poor. The inactive poor also include working age population, and especially women. I]ILO 2002, "Decent Work inthe informal economy" 13 33. High inactivity rates, especially among women, in poor households lead to higher dependency rates. Poor households have a higher share of dependents relative to working members, reflecting lower utilization o f their labor resources. This is in large part due to the much lower labor force participation rates among poor women: while poor males are 8% less likely to participate inthe labor market, poor women have 25% lower participation rates compared to the non-poor. Higher female inactivity is linked primarily to lower educational attainment o f women, particularly o f technical and higher education, but may also be due to cultural factors and the presenceof very young children. C. Health andPoverty 34. Despite some achievements in health indicators, the levels and trendsof some health and nutrition outcomes are of concern. Low children's nutritional status could have serious long-term development consequences. The latest household survey results available unfortunately coincide with the lowest point o f the transition recession, in 1996. At that time, malnutrition among children was high (3 1% of children under 3 years o f age were stunted and 19% were under-weight in 1996) '*.As inother republics of the Former Soviet Union, the infant mortality rate based on survey data i s much higher (76% higher) than reported in administrative data for the same year13.Anemia rates among very young children and women o f reproductive age are also among the highest inthe region. There have been recent improvements in preventable infectious diseases. However some infectious diseases are a major concern and remain the most important cause of death and morbidity among children. Tuberculosis incidence has increased by nearly 30% between 1990 and 1998 and AIDS is rising e~ponentially'~ 35. The poor are the most affected by inadequate health status. The incidence o f malnutrition and infectious diseases tend to be higher in poorer and more rural regions, although this correlation between poverty rates in regions and health outcomes is not perfect (Figure 4). Karakalpakstan is particularly vulnerable, and the incidence of anemia and tuberculosis in this region is the highest in the country. Malnutrition is high in Namangan and Surkhandarya, among other regions. Poor health and nutrition status reflects differences inaccess and quality of services (including water and sanitation, health services), but also economic barriers (low incomes, highformal and informal costs of care) and cultural influences. 36. But many non-poor but chronically ill are also at risk of falling into poverty. While the burden of out of pocket payments i s disproportionately high for the poor, many non-poor households are at risk of falling into poverty as a result of substantial health expenditures occurring due to catastrophic or chronic illnesses. For households with chronically ill or disabled members, average health care expenditures as a proportion o f food expenditures were two and a half times that for the average household and constituted between half to 60% o f food consumption for the second and third income quintiles. Faced with highcosts, many poor and near poor employ counterproductive coping strategies -they deplete savings, borrow money, sell assets, reduce expenses including on food, and delay or avoid seeking care. 12 Fromthe 1996DemographicandHealthSurvey. Although not availablefor citationwhen this reportwas going to press, preliminary resultsfrom the 2002Demographicand HealthSurvey indicate significant improvementsin indicatorso f stunting, wasting and underweightchildrenunder 3 years of age, but not in anemia rates. 15 The discrepanciesin survey and administrativedata are not uniqueto Uzbekistanbut are found in other former Soviet Union Republics:they arise in part from differencesin definition, but largely from differencesin design(household basedversus registry or institution based) and implementationof the survey. j 4Officially there are less than 2,000 cases, but recent UNAIDS statisticssuggest that the numbercould be morethan 6,000. 14 Figure 4: Nutritional outcomes are worse in rural areas andparticular regions ~ - ~. ~~~~ Regional incidence of underweight children 40 30 20 x) 0 D. EducationandPoverty 37. Some educational achievementsare at risk. Official data show impressive gains in basic education enrollments since the mid-l990's, with the net enrollment ratio at 95% in 2000-01. At all other levels, however, the educational achievements inherited at transition now appear to be at risk. In2000/01, absolute enrollments at pre-schools and higher education institutions were only about 50% o f 1991/92 levels. The substantially lower enrollment ratios at all levels other than basic education suggest that a much smaller fraction o f the school age population are now covered by the education system at pre-school, secondary and higher levels. The fall in higher education enrollments in Uzbekistan is in contrast with other transition countries. It is important to note, however, that very recent years have seen important reversals o f this negativetrend. 38. There are disparities in attendance between regions, income groups and gender. The poor have lower enrollment rates particularly at teknikum and higher levels o f education. Significant disparities also remain between men and women at higher levels of education, particularly university education: female students constitute only a third o f enrollments. Enrollment rates at these levels are highest in Tashkent City and Novoi, and lowest in Karakalpakastan, Surkhandarya, and Kashkadarya. As in the case o f health outcomes, regional differences in enrollment reflect differences in access and quality, but also economic barriers (low incomes, highcosts), and cultural influences. 39. However education does not protect completely from poverty. Consistent with evidence from CIS countries, poverty rates are not different across households heads with basic and secondary education. More than half o f the poor have householdheads who have completed secondary education. However, higher levels of educational attainment are strongly correlated with lower levels of poverv. Families where the head of the household has higher education are half as likely to be poor as those where the head has nine or fewer years of education (Figure 5). These findings are indicative o f a transition labor market where earnings differences do not yet fully reflect productivity differences. Only higher education graduates command a significant wage premiuminthe labor market. The lack of a premiumto secondary education relative to basic education inthe non-state sector, i s incontrast to countries in Central and Eastern Europe and the Baltics, and may also reflect insufficient demand for the skills of its graduates. 15 Figure 5: Education does notprotect completelyfrom Poverty ~~ Poverty and Education Level of HouseholdHead 60 - None to fouryears Fwe to nmeyears Fullsecondary Incomplete Technkum CompleteTecWSome Hghereducatm higher ShareofPopuktnn I l S h a r e o f P o o r A h c l d e n c e o f P o v e r t y .-.-. - -PovertyIncidence-Natlonal , 111. KEY CHALLENGESTO IMPROVING LIVINGSTANDARDS 40. Low living standards are thus linked to insufficient job opportunities, low and uncertain incomes, particularly in rural areas, low returns to education, and lower capabilities (health and education) o f the poor. What are the main factors that contribute to these outcomes? What is the role played by Government policies and programs? How effective are the Government's extensive poverty alleviation programs (i.e. public social protection programs) inprotecting the poor? 1. Constraintsto Job and IncomeGrowth A. LimitedPrivateSector 41. The SME sector remains limited in Uzbekistan, despite recent Government reforms. Unlike the more advanced transition countries, the private and small firm sector have not yet emerged as significant providers o f employment opportunities. The Government initiated, in2000, several reforms to encourage the S M E sector. These include the streamlining of business registration procedures, elimination of exchange surrender re uirements, reductions in business inspections and simplification of tax and reporting requirements . The S M E sector appears to have responded to these reforms, with an increase in the official 8 numbero f registered SMEs. Nevertheless, in2001 formal SMEs accounted for only 15 percent o fGDP and 9 percent o f total employment while individual entrepreneurs, many of them informal, and small dehqun (household plot) farmers accounted for 19 percent o f GDP, as much as 41 percent of total employment16. These latter groups also accounted for most o f the growth in the S M E sector, which may be indicative o f the increasing informal sector. 42. Theprivate and SME sectorfaces an unfavorable investmentclimate. The provision of privileges to the S M E sector in isolation o f an overall improvement to the environment for private sector development has l5Inthe second half of 2002 and early 2003 some reversals, which raise concern, took place. These include a clamp-down on individual traders, sharp tightening of import procedures, attempts to centralize wholesale trade, a threat to re-privatize some trading enterprises and more pervasivecontrol over financial transactions of enterprises. l6Source: World Bank, Country Economic Memorandum, 2003. 16 limited impact on firm entry and development, and can provide small firms with the incentive to remain small or large firms to fragment. The following are top constraints to the environment for private sector de~elopment:'~ 43. Inflation, restrictions on access toforeign exchange, and related trade restrictions are the most significant barriers identified by SMEs and private firms in Uzbekistan (much more so than in other countries)". Although recent reforms have reducedthe surrender requirements of SMEs, all other exports are still subject to heavy foreign exchange surrender requirements, and access to foreign exchange remains limited by import contract registration requirements, and other formal and informal restrictions. Moreover, despite recent steps towards unification o f its multiple exchange rates, Uzbekistan still had two legal exchange rates and a gap o f some 10-15% between the officiaVover the counter and curb market rate remainedinApril 2003. 44. Major financing constraints include high cost of funds and difficulties in accessing bank loans. These constraints can be directly related to macroeconomic and structural distortions which raise the cost of funds to the private sector. They can also be attributed to the significant state intervention in the banking system which undermines its credibility and diminishes its ability to channel resources to productive investments. Limitations on access to - and usage o f - cash, and underdeveloped legal institutions to support credit markets are additional reasonsfor financing constraints, especially for SMEs. 45. Soft budget constraints to state enterprises contribute to an unevenplayingfield for SMEs. Large enterprises benefit from substantial explicit and implicit subsidies, payment arrears, budget offsets, tax exemptions, publicly guaranteed loans, and preferential access to inputs such as foreign exchange and credit. Excessive state intervention also occurs through controls in production, prices, marketing, and allocation of inputs(exercised through Productionand Trade Associations, PTAs), highstate ownership in key sectors, and limitedeffectiveness o f the Anti-Monopoly Committee. burdensome. Although they are gradually being reduced, taxes are still highI?)requirements 46. Taxes and tax administration practices are costly, and regulato are still and are accompanied by various tax exemptions and privileges. These taxes, their collection through the banking system, and onerous reporting requirements combine with a narrowing tax base (as a result o f the proliferation o f tax breaks to strategic enterprises and, recently, small firms) to contribute to a vicious cycle of increasing informality and hightaxes. The tax exemptions and privileges reduce transparency and create disincentives for firm growth. Although there has been progress particularly on streamlining entry; licensing requirements and inspections are still dramatically higher than inOECD countries. 47. Corruption and informal payments constitute a heavier burdenfor SMEs. Heavy state intervention and discretion in the tax and regulatory regimes provide officials with opportunities for direct contact with firms, greater discretion, and therefore more possibilities to extract informal payments from entrepreneurs. Within transition countries, Uzbekistan ranks third in the frequency o f informal payments, behindAzerbaijan and Romania. SMEs are disproportionately affected: 66% o f small firms in Uzbekistan report making bribes compared to 23% for large firms or 43% for all firms. Poor regulatory quality and limited transparency are important contributors to poor governance and lower accountability of state institutions. Institutions to support access to information are weak and the media is almost entirely state owned and restricted by tight l7 These findings are drawn from two surveys: a survey of 125 firms in 1999 (conducted by the EBRD and the World Bank), and surveys of 4000 entrepreneurs and managersof SMEsas well as private and dehqan farms conducted by the Center for Public Opinion rjtimoiy Fikr), on behalfof the Government of Uzbekistanin 2000 and 2001. I* For a more detailed discussionof Uzbekistan's macroeconomic and structural policies see World Bank, Country Economic Memorandum, 2003. l 9They include a profit tax of 20%for 2003, value added tax (VAT) of 20% and a complex array of other taxes (local, property, etc.). 17 licensing and defamation laws. As in other transition countries, civil society groups and alternative business associations are still nascent, and therefore play a limited role in increasing the voice and participation of businesses inpolicy formulation and in enhancing accountability of state institutions. 48. Rigidities in some labor market institutions may raise the cost of labor in theformal sector and disproportionatelyaffect labor market outcomesfor marginal workers. Flexible labor market institutions are essential for job creation and labor market efficiency. Uzbekistan has relatively flexible labor market legislation. However, the use of the Current Tariff Scheme for wage setting and high taxes on labor (including payroll taxes2') induce rigidities in the labor market by raising the cost of formal employment, particularly for youth, unskilledworkers and workers in poorer regions. Other constraints to a flexible labor market are possible barriers to mobility arising from policies within the labor market (e.g. propiska) or outside (e.g. housingmarket). These may contribute to regional differences in unemployment rates. B. Constraintsto RuralIncomes 49. Agricultural productivity growth and incomes are constrained by significant state intervention in oulput markets,particularly in the cotton and grain sectors. The state intervenes through pricing, exchange rate and procurement policies, and mandatory cropping patterns. While the implicit taxation of agriculture has been reduced, mainly by a narrowing in the gap between the prices paid for state procurement and internationalprices, and represents considerable progress, it still remains significant. Currently, taxation takes place primarily through exchange rate policies. A sizeable amount o f tax revenues are returned to the agriculture sector in the form o f subsidies, the most substantial being irrigation subsidies and periodic debt write-offsa21 However, agricultural enterprises, particularly those involved in cotton, still regularly incur losses. Price distortions combined with mandatory cropping plans leadto inefficient output mix and landuse. 50. State interventions in input markets reduces the quantity and quality of inputs available to all farmers, with adverse impact on agricultural productivity. Farmers have limited options (outside o f state input suppliers) for input purchases, and limited access to capital. The provision of working capital through the Fund for State Agricultural Purchases (through accounting adjustments between input suppliers), limits the access o f farmers to cash and precludes the development o f private sector input suppliers.Direct provision o f water by the state, poor maintenance o f the irrigation and drainage system, and subsidization through inadequate water charges contribute to inefficient and wasteful water use, reducing agricultural productivity. Italso increasesthevulnerability ofthe rural populationto periodic droughts. Farm Enterprises Private Farms Dehqan Farms Share of Total Value of Cror,OutDut 48% 9% 43% Share of Total Land 72% 16% 12% Ratio 0.67 0.56 3.58 5 1. The ongoingprocess of transformation of shirkats into largeprivatefarms raises equity concerns. Inthe current phase of landreform and restructuring, unprofitable shirkats are beingliquidated, and their land i s being used for the creation of new large private farms. After an initial expansion in household (dehqan) *' Social security taxes were 39.7% in2003, among the highest inthe region, and significantly higher thanthe average for EUand OECD countries. Payrolltaxes payableto the PensionFundonly include a 35% payroll tax from employers and a 2.5% mandatory *'insurance fee levied on employees' wages. Implicit taxes and subsidies balancedoff in 2001, and when debt write-offs are taken into account there was a net transfer of resources to the sector 18 plots in 1991, the acreage of land in the household sector has expanded very slowly. Thus poor households, who are less likely to be awarded land for private farming, are less able to expand their land assets. Although access to household plots is widespread, the average size o f dehqan plots is only about 0.16 hectares. 52. The dynamism of the dehqanfarm sector suggests that agriculture could be moreproductive with improvementsin incentives and increasedflexibility in production. Table 1 shows that dehqan farms are the most efficient: their contribution to output i s three and a half times greater than the share of landused by such farms. Improved incentives (higher prices and incomes), as well as flexibility to set their mix o f output are key. Moreover, the smaller size of their land holding i s not necessarily associated with lower land productivity, at least at the current allocations o f land between farm types. These plots also remain important source o f income for poor households, thus playing an important role inrural poverty reduction. 2. Challengesto Improving Capabilities A. Health Care 53. To ensure the continuedprovision of health care, particularly for rural and poor populations, the Government has been implementing, a comprehensive health care reform program Public expenditures on health care fell from about 6% of GDP prior to independence to about 2.6% o f GDP in 2000. As a result, health services suffered considerably, particularly in the initial years o f transition. As a response, the new Government program, aimed at improving efficiency, emphasizes the development of primary health care, reduced reliance on higher end and specialized care, introduction o f formal charges for inpatient and tertiary care, and gradual increase in private provision of health care. Implementation progress has been mixed, with faster pace o f development o f emergency care facilities and slower progress on rationalization o f higher end facilities. Early reforms to strengthenprimary care facilities benefited better off regions, althoughthis i s now changing. 54. Private paymentsfor health Gformal and informal) which have increased as publicJinancing has declined,constitutea significant burdenfor thepoor. The fall inpublic payments has been accompanied by an increase in the incidence o f private charges for health care (both formal and informal). Taking these into account, total health expenditures equal to 64% o f GDP in 2000. Informal payments are significant-more than two thirds of health users interviewed in a qualitative survey reported paying for health services informally. Informalpayments were more likely (and higher) for complex services for which formal payment i s required, compared to those widely recognized as free (such as immunization). While better off- populations are able to by-pass referral systems and access higher end care, the poor are less able to make private payments. They are thus less able to purchase higher quality care, or benefit from hospital care, although private payments are disproportionately higher as a share o f income for the poor. 55. Public expenditures in health do notfavor the poor. Most rural poor frequent less costly public primary care clinics, as well as rural clinics and hospitals, while higher income urban groups seek more expensive publicly financed hospital care intowns and large cities. Since hospital expenditures constitute the majority of public expenditures on health care, the well off and urban populations consume the majority of public health resources (Figure 6). The high level of funding required by the new emergency facilities may further crowd out public budget resources for primary health care and outpatient services for the poor. 19 __ Figure 6: Public Expenditures on~-~ Favor the urban and better off Impatient Care ~ I Incidence of Public Expenditures on In-Patient Care I1 70.0 600 ~ ,I 500 * 400 n 300 I 20 0 100 I 0 0 56. Policy provisions to allay costs of higher formal charges do not adequately protect the poor. To protect vulnerable groups, many legal provisions have been introduced that exempt certain individuals from payments for medical services, pharmaceuticals and equipment. However, consumers may not always understand entitlements and enforcement i s uneven. Second, the exempt groups as specified are defined on the basis o f broad categories and do not correlate very strongly with poverty status. This may partly explain the highburden of private expenditures for the poor despite formal efforts at protection. 57. The emerging model of private provision of health care raises additional equity concerns. Privatization o f health care provision can be important for improvingthe quality and efficiency o f health care services. However, the emerging model for privatizationof health care raises the following concerns. 0 Informal provision o f private health care by doctors employed at public facilities seems to be emerging as a significant mode o f private delivery. The majority of private providers are either unlicensed, do not have formal contracts with or do not reimbursethe public facilities for privateprovision. The privatization of care envisions specific facilities o f specialized care being privatized. Without adequate protection to protect access to these services, privatization (once completed) may result inthe poor (particularly in small urban areas) beingdeprived of these services. 58. Informal payments and practices in the health care system are associated with lack of clarity in regulation, and underdevelopedmechanismsto ensure accountability of health sector institutions. Informal payments and practices divert resources away from the system, reducing resources available for improving public services, and underminingpublic service delivery. Lack o f clarity and understanding by doctors and patients on who should receive free services and from which facilities, or what constitutes paid services, or on setting tariffs for paid services is an important contributor to informalpayments. The lack o f formal oversight mechanisms to monitor provider charges or performance and to enforce policies contributes to low accountability of health service institutions. Like other transition countries, civil society organizations are still nascent and play a limited role in increasing the voice and participation of health users. Mahallahs play an advocacy role in some areas, buttend not to be substantially involved in health care. 20 Figure 7: Public expenditureson educationare slightly regressive Incidenceon Public EducationExpenditures 0 20 40 60 80 100 Cumulative Share of Population (%) +-Primary - a GeneralSecondary --A-PTU/SPTU - - -X- -Tekhnikum +University - +Total B. Education 59. Although smaller than other CIS countries, the decline in public expenditures on education were accompanied by decentralization of financing and likely reduced the quality of education in poorer communities. The expenditures on education declined less than in other CIS countries from about 10% of GDP to 6.7% between 1990 and 2000, reflecting the Government's commitment to education. Still, the fall in resources was significant especially in the first half of the 1990's. Declining public resources were initially accompanied by the decentralization o f financing and management for most educational programs to oblast governments. While these reforms are intended to increase the flexibility and reduce costs of the public education system, they may have led to a disproportionate decline inquality o f education in communities that were unable to supplementpublic financing with additional private resources. 60. Private paymentsfor education (bothformal and informal) have also increased, and constitute a disproportionate burdenfor thepoor. Increased formal payments (e.g. textbooks are now only provided free o f cost to grade 1, formal cost recovery has been introduced in specialized secondary and higher educational institutions) also accompanied the reduction and decentralization of public finances in education. However, these higher payments (e.g. an average annual tuition fee for higher education is now 300,000 soms), are burdensome for the poor. Private education spending constitutes 16% o f food expenditures for poor households, compared to 9% for the non-poor. Informal costs also appear to be important, particularly in highereducation, where over a fifth of all households reportedmaking unofficial payments or providing gifts to staff. Low or falling incomes, combined with increased private costs and lower perceived quality or relevance (for example lack o f employment opportunities for young graduates) may be important reasons for declining education enrollments, particularly for the poor. 61. The overall distribution of pubiic education expenditures slightly favors the non-poor (is slightly regressive) (Figure 7). While primary education expenditures are distributed equally across all income groups, public expenditures on technical and university education disproportionately benefit the better off households. This is not surprising: public subsidies per child are larger, and disparities in participation of 21 children from higher (vs. lower) income households is greater at higher (vs. lower) levels of education. Moreover, public subsidies to cover the high formal costs of higher education are provided on the basis of merit, without consideration o f means. 62. Government efforts to improve the quality and market relevance of public education serve some crucial needsparticularly at basic education, but some other aspectspose risks. In 1997, the government launched a major program to improve the quality and relevance of the education system. Many crucial inputs are beingsupplied, starting from the base of the education pyramid, addressing some key needs of the system. However, the program's heavy focus on upgrading secondary and higher education (for example the substantial investment in infrastructure for establishing the large network o f professional secondary schools), and the associated recurrent cost requirements, risks drawing resources away from basic education. There is also a risk that the emphasis on occupation specific training at the secondary and higher levels may prove too rigid for the needs o f a market economy, while the channeling o f a small fraction o f secondary education students into academic lycea attached to universities may prove to be elitist and increase disparities in opportunities. 3. Challenges to Providing Effective Social Protection 63. Uzbekistan's social protection programs are extensive, for its level of income, but expenditures have declined in recent years. Uzbekistan has in place an extensive social protection program to protect individuals against life cycle risks and poverty. As in OECD and other transition countries, the main programs are pensions, unemployment benefits, social assistance (including child allowances and poverty benefits), and privileges. In 2000, Uzbekistan spent over 8.24% o f GDP on pension programs, employment programs and means-tested poverty benefits. However, spending levels have declined over time, reducing average benefit levels and number of beneficiaries for some social programs, particularly poverty benefits. Figure 8: Child allowances are the best targeted,housingprivileges werepoorly targeted Targeting Efficiency of Utility Expenditures Targeting Efficiency of Social Assistance Programs 100 90 80 70 60 50 40 30 20 10 0 -+ - - 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 -Equality PC FoodConsumption -Equality PCFoodConsumptton -t-Rent - - - 60 70 80 90 100 +Electricity +Mahalla-Poor Mahalla-Child ---31tGas 1-X- Child Benfit under 2 64. Pensions provide old-age security to many elderly but at high costs to formal employment. The pension system provides extensive coverage o f the elderly and is the most generous o f all social protection systems. It has had a substantial impact on poverty reduction among pensioners and their households. However, the fiscal and incentive costs of this system are substantial: high payroll taxes discourage formal 22 employment, while incentives for early retirement likely reduce labor supply, particularly for women. Higher dependency ratios in the future will adversely impact the future sustainability of the system 65. Unemployment benefits have limited impact onprotecting the unemployed while most expenditures go to costly active labor market programs. The coverage of unemployment programs is very low (only 8% of unemployed register and only a fifth o fthese draw unemployment benefits), and much lower than coverage in most OECD and CEE countries. More than halfo f employment fund expenditures are devoted to costly active labor market support (particularly subsidized micro-credit schemes). International experience indicate that direct job-creation schemes have limited impact on creation of private sector employment opportunities, while programs such as job-search assistance and counseling can be very effective in a market economy. Although administrative data suggest high success rate associated with active employment programs, the best way to gauge success o f active labor market programs (ALMPs) is to undertake impact evaluations. 66. The Government has achieved a measure of success in targeting child allowances and poverty benefits, but there is scope to improve targeting efficiency across and within regions. Child allowances delivered in a decentralized fashion through Mahallahs cover a quarter o f the poorest quintile, while the coverage o f the Mahallah low income benefit i s much lower. Child allowances also have better targeting efficiency, but errors of inclusion and exclusion can be improved for both programs (Figure 8). Targeting of these programs acrossregions can also be significantly improved. 67. Social privileges provided generous benefits but were poorly targeted. Just under 1% o f GDP (higher than expenditure on poverty benefits in 2001) was spent until recently on social privileges to education and health care workers, primarily in the form o f discounts on utility tariffs and maintenance fees. These subsidies were quite generous and intended, by design, to reward particular occupations, and not to protect the poor. Not surprisingly, these subsidies were poorly targeted (Figure 8), and the better off and urban residents benefitdisproportionately. Ina significant policy initiative, more or less at the time this report was going to press, these subsidies have now been largely abolished. IV. POLICYPRIORITIES FORIMPROVINGLIVING STANDARDS 68. Five priority areas are thus suggested for a strategy to improve living standards in Uzbekistan: (1) promoting income and employment opportunities through economic growth (2) improving capabilities, particularly o f the poor; (3) providing security through social protection; (4) building inclusive institutions and facilitating empowerment; and (5) strengthening poverty monitoring. The strategy should place particular emphasis inrural areas and poorer or needier regions. Key actions are discussedbelow. 1. PromotingOpportunities throughGrowth 69. The primary pillar o f a strategy to improve living standards in Uzbekistan is sustained private sector led economic growth: through reforms to encourage growth and employment generation by the private sector (including SMEs), and to increase agricultural growth. These reforms will benefit the entire population, including the poor, but some groups may be adversely affected by the reforms inthe short term. These groups will need to be protected through well targeted social protection programs. Sustained growth and poverty reduction will also require attention to flexible labor market policies, that allow labor to be allocated to its most productive use. A. PromotingPrivateSector and SME development 70. Increase the demand for labor through an improved environmentfor private sector and SME development. Experience from high growth transition countries has shown this sector is the main generator ofjobs and productivity growth for the economy. 23 a Promote macro-economic stability and proceed with fundamental structural reforms. Key reforms include the liberalization o f the foreign exchange and trade regime (in conjunction with tight fiscal and monetary policies to contain inflation), budget management and financial sector reforms, further reduction in state intervention inproduction, prices and marketing, and continued progress inprivatization. Improve the business and competitive environment for private firms and SMEs. The main reforms include the imposition of hard budget constraints on state owned enterprises to create a level playing field, improving the legal and institutional framework for access to credit, tax and tax administration reform, and further reduction in licensing and regulatory requirements. Box 3 shows that privatization has been associated with large gains in productivity profitability, efficiency and output, albeit with some employment costs inthe short term. Box 3: In Latin America, Privatization yielded many benefits,but had some short term costs... Privatization has done much for Latin America. Studies show that thanks to privatization, several Latin American countries have enjoyed greater investment, profitability, efficiency, and output. A series of studies analyzed costs and benefits of privatization in Argentina, Bolivia, Brazil, Chile, Colombia, Mexico and Peru. Figure 1 summarizes the empirical results of the impact of privatization on four major indicators: profitability, operational efficiency, output and employment. Inall four countries profitability, operational efficiency and output were up while short-run employment suffered but the magnitude of the impact varied. Interms of profitability, except for Colombia, Chile and to a lesser extentBrazil, public companies prior to privatization suffered huge lossesthat were financed by the state at the expenseof a ballooning fiscal deficit. Given greater profitability, it is not surprising that operational efficiency was vastly improved and output was also dramatically up. The costs of privatization were felt through a fall in employment in affected firms. However, this refers only to short-run, direct, stable employment. More recent studies indicate that post-privatization employment is increasing, either through re-hiring of previously laid-off workers or new hires. Moreover, when new indirect jobs generated inprivatized sectors are considered, the employment picture is muchbrighter. Source: Latin American Economic Policies. Second Quarter, 2002, B. IncreasingAgricultural Growth and ReducingRuralPoverty 71. Particular attention should also be paid to enhancing non-farm agricultural opportunities in rural areas, through the measuresdiscussed above. 72. Reduce distortions through pricing and procurement, production, and input policies. Priority actions for improving agricultural productivity and incomes, which would have a significant impact on rural poverty, are: (a) complete the process of alignment of cotton and grain prices to world market prices through 24 foreign exchange rate liberalization- as discussed below, this will significantly benefit the rural poor; (b) reduce procurement quotas and eliminate mandatory sown crop areas, accompanied by encouraging private sector participation in product markets and agricultural processing; and (c) reduce input subsidies and encourage private sector provision of inputs.The recent experience o f the Kyrgyz Republic shows that major agricultural reforms in the area of land policy, the incentive framework, credit, irrigation and agricultural services, leads to significant agricultural growth and reduction inrural poverty. The country realizeda growth rate of 6% per year, which lead to a steady fall in rural poverty- at an average annual rate o f 8% per annum The experience inChina also supports the importance of agriculturalreforms inalleviating poverty (Box4). 73. Increase the equity of land reforms. In the next stage o f transformation of shirkats, consideration should be given to reducing the legal distinction in land size between private and dehqan farms, and permitting the expansion of household plots for shirkhat members (above the current 0.16 hectares). These changes should be accompanied by the creation of a legal and institutional framework for leasing and consolidation o f land use rights, to facilitate the use of land where its marginal productivity is highest. Author Dependent variable studied Period (Yo) Method Lin (1987) Grossvalue of agriculturaloutput 1980-83 62 Productionfunction McMillan, Whally, andGrossvalue of agriculturaloutput 1978-84 51.8 Denilson-Solow Zhu (1992) accounting Fan(1991) Gross value of agriculturaloutput 1965-85 56 Productionfrontier Carter and Zhong Grain yield 1979-86 19.5 Grainyield function Lin (1992) Grossvalue of crops 1970-87 46.9 Productionfunction Huangand Rozelle Rice 1975-90 35.6 Ricevield function A more recent and detailed study on the effect of agriculturalreform on agricultural productivity, for the case of grain productionacross five agro-climatic regions in China over the period of 1980-90 shows that productionefficiency gains from economic reforms between 1980 and 1985, on average contributed 38% to the growth in grain production in China. The productionefficiency gainswere sustaineduntil 1987 andthen fell in 1988 and 1989 to 24% of contributionto the annualgrowth rate in grain production. Source: "Reforms, the weather, andproductivitygrowth in China's grainsector". Zhang, Bin.;Carter, Colin A. American Journal ofAariculfuru1 Economics v. 79 no4 (Nov. 1997)p. 1266-17 C. The DistributionalImpact of Reforms 74. Foreign exchange liberalization will have a generally beneficial impact on rural incomes; however as with other actions such as privatization, reductions in price controls, or improvements in the effectiveness o f the anti-monopoly committee, it could increase the cost o f living through higher prices for some goods, including possibly essential commodities consumed by the poor. The evidence for Uzbekistan i s that while 25 both urban and rural households will face higher (food) cost o f living increasesas a result of foreign exchange liberalization, the relative initial increase in cost of living will be higher for the urban population (roughly equally for all quintiles), and for the poorer quintiles in rural areas. Over time, however, individuals will adjust their consumptionpatterns and consume cheaper substitutes where possible. 75. Aligning cotton and grain prices with world market prices would immediately increase incomes for a large fraction of the rural population. While the increase inabsolute terms is expected to be roughly equally distributedacross all income quintiles, poorer farmers (who derive a larger share of their income from cotton and grain), will benefit more inrelative terms. Figure 9 demonstrates this for cotton farmers. A similar result i s found for grain farmers, although the magnitudes of the impact are smaller. In the short term, however, some groups, for example net consumers o f wheat (in urban areas, but also in rural areas), may face higher prices. Figure 9: Increasing cottonprices to world market levels improves income of all andpoorfarmers ' Welfare Impact of Increasing Cotton Prices to World Market Levels I Percent I Income Change T 140% (thousands 2000 - 120% Change o f soum) - 100% - t 80% 5 ;;; 60% - 0% 1 2 3 4 5 ' Welfare Quinule - - - I I n c o m e Change--Cotton Farmers (thousands o f Soums) -YO Change in Income ~- 76. Enterprise Restructuring that would result from the imposition of hard budget constraints, the removal of explicit and implicit subsidies to enterprises, or foreign exchange rate liberalization can be expected to be associated with employment losses, particularly in the short term. Many countries in Central and Eastern Europe experienced short term employment losses as a result of economic restructuring. However, countries that restructured more rapidly and moved quickly to reduce institutional barriers to job creation, such as Estonia, realized far greater gains than those who took a slower, more interventionist approach, e.g. Slovenia (see Box 5) .Most countries protected the most affected workers by targeted social protection programs. D. EnsuringFlexibility of the Labor Market 77. Reform labor market policies and institutions to reduce labor costs and increase labor market flexibility through a reduced role of the Common Tariff Scheme for wage setting and the reduction o f non- wage labor costs (for example, payroll taxes, in the context of broader social protection reform). Implementingthe recent legislation for the abolition of the tariff grid for wage setting in all enterprises except budgetary institutions as well as reducing non-wage labor costs will help ensure that reforms to improve the investment climate are accompanied by adequate supply response for employment, particularly inpoorer, low productivity areas and for marginal workers. Examining and reducing barriers to labor mobility, including potentially, the use of thepropisku to control entry into Tashkent, could facilitate geographical movement of labor in response to new opportunities. 26 78. Ensure that the laborforce has the skills that are demanded by the new private sector. This will require reforms inthe education system so that new labor market entrants have the skills that will help them meet the emerging and changing needs of a market economy (see education section below). Workers laid off as a result o f restructuring can be re-trained through targeted training programs, provided with substantial involvement o f private sector firms (see social protection section below). Targeted actions to encourage and increase their participation o f women in higher levels of education could reduce inactivity rates among women. Box 5. Estonia and Slovenia: Policy choices and worker reallocations during transition The cases of Estonia and Slovenia offer important insights on reform options to encourage employment and earnings growth in the market economy. The two countries saw very different pattems of labor market adjustment interms of labor market stocks and flows across different labor market states during transition. Estonia followed a liberal approach, with few barriers to labor market dislocations or newjob creation, meager support of the unemployed, no effective wage floor, low taxation of labor, and privatization methods (primarily direct sales of assets to strategic investors)that strengthenedcorporate govemance and thus encouragedlabor shedding. Incontrast, Sloveniaadopted a muchmore cautious, interventionistapproach, with significant barriers to job dislocation (including subsidies to prevent layoffs), generous support for unemployed workers, rigid wage setting mechanism,andmostly insider privatization thathinderedreductions inemployment. Estonian reform generated large worker and job flows that have facilitated intense labor reallocation across sectors including high job destruction but also larger job creation rates including the creation of many productive jobs while simultaneouslyallowing the destructionof unproductivejobs. Directjob-to-job transitions strongly increased, morethan doubling their pre-transitionrate, and throughout the transition about half of the workers who lostjobs have been able to transfer to a new job without an interveningperiodof unemployment. Incomparison, amore"sclerotic" Slovenianlabormarketproducedmore segmentationandworse labormarketoutcomes especially for marginal groups (younger workers, ethnic minorities). Although the job destruction rates in Slovenia even at the height of restructuringremainedbelow 60 percent of the maximumEstonianjob destructionrate, thejob creationrates in Slovenia were many times below the ratesinEstonia, implyingmuchhighernet employment generation in Estonia. Sources: M. Vodopivec. Worker Reallocation During Estonia`s Transition to Market: How Efficient and How Equitable? The WorldBank, SP DiscussionPaper,No. 0018,July, 2000. 2. ImprovingCapabilitiesthroughInvestmentsinHealthand Education 79. To ensure that individuals, and especially the poor, are able to avail o f the opportunities provided through growth, a poverty reduction strategy for Uzbekistan must focus on building their capabilities, particularly their health and education. A. ImprovingQuality and Access to Education 80. Ensure that the education system provides relevant skills that enhance the ability o f individuals to findjobs and earn higher incomes. At all levels, attention should be paid to establishing a continuous process o f evaluation o f the skills provided through the system, through standardized national tests and participation in international assessments22,and making continuous adjustments and improvements. Specific measures include: At the primary level, continue efforts initiated under the National Program to improve quality of education, through ongoing reforms to ensure the provision o f crucial inputs, modernization o f curricula, and 22 Studentsfrom Uzbekistancurrently participate in internationalcontestsand Olympiads. These however, are for the best students. Participationin internationaltests such as PISA, TIMSS, etc. provideinformation onthe performance of the majority of children in the system. 27 introduction o f student centered teaching practices. Improve teacher incentives and link them with teaching loads and improved educational outcomes. 0 At the secondary level, reevaluate the current plan to restructure secondary schools into professional colleges and academic lycea, to ensure that more general, academic skills, which are proving to be most in demand inmarket economies, are providedto the majority of students at this level. At the tertiary level, improve the understanding of falling university enrollments, and examine options for reversing this trend (e.g. encourage private sector provision, introduce subsidies on the basis of merit and means). Examine andundertake actions to increase enrollments o f women inteknikum and university. 81. Improve the targeting of public education resources to address disparities in attendance and quality between poor and non-poor and between regions. Introduce flexible financing mechanisms to enable directing o f resources on the basis of need. Wherever possible, encourage local solutions by communities to increase attendance and quality. 0 Strengthen targeted efSorts to increase attendance by the poor. Continue and strengthen the existing programs of targeted provision of textbooks and educational materials for basic education, pilot other approaches (such as conditional transfers, scholarships- income and merit based, school lunches) that have had some success inother countries. 0 Adopt targeted approaches to early childhood education to address the concerns o f late enrollments and to improve performance in basic education among vulnerable groups, building on existing innovative approaches. B. ProtectingHealthStatus andPreventingImpoverishment 82. Reduce the incidence of some infectious diseases and continue efforts to improve nutritional outcomes, especially among women and children. Many required interventions lie outside the health sector, and concerted efforts to work across the traditional boundaries of Government ministries will be required. Recommended actions are: (a) continue the program of restoration of primary health care services with particular emphasis on poorer rural regions; (b) place additional emphasis on improvements in water and sanitation services, particularly in rural areas and poorer regions; (c) continue efforts that have been initiated to strengthen programs for addressing TB and AIDS; (d) continue efforts on maternal and child health and monitoring; (e) encourage community based information and behavioral change programs; and (f) strengthen programs for fortification of foods (e.g. iron, iodine), etc. 83. Improve the targeting of public health care resources to thepoor. Actions include: Increasing relative allocations to primary health care through acceleration o f the program of restructuringthe hospital sector (but with greater consultation o f local stakeholders), and introducing flexible financing mechanisms to permitresource allocations to poorer and more needy regions and sub-regions; 0 Reviewheplace existing specifications of exempt groups with alternative approaches to targeting that are more closely related with poverty status. The existing package of exemptions (from care at all levels) could be delivered to the poor households thus identified, but it should be monitored carefully to ensure that there are no gaps inaccess to key services by the poor, particularly as privatization proceeds. Additional efforts are also neededto enforce these exemptions and protect the poor from informal charges (see below). 84. Protect the non-poorfrom falling into poverty as a result of costs of catastrophic or chronic care. New private insurance is envisioned, which could protect the non-poor from costs associated with catastrophic or chronic care. However, international experience has shown that these schemes need to be designed and regulated carefully. Alternatively, limited public subsidies, targeted in some way could be considered to protect the vulnerable groups from falling into poverty. 28 85. Foster development of private sector, but with greater protection to consumers. As the formal private sector develops, consider ways to foster its development while at the same time protecting consumers and monitoring quality: regulations and accreditation standards need to be further developed and monitoring mechanisms need to be strengthened. The emphasis should be on setting incentives and maintaining quality rather than imposing a burden on private providers. 3. ProvidingEffective Security through SocialProtectionPrograms 86. The Government has initiated reforms to strengthen the social protection system and improve its efficiency and effectiveness. 87. Provide effective old age protection through thepension system. Specific actions, which may lead to an effective security system that is financially viable and which has lower costs to formal employment include: (i)introduce parametric changes to the pension system (e.g. increase and equalize retirement ages and years of service requirements, and phase out pension privileges). Ifphased in systematically, these could help reduce fiscal pressures inthe system, without compromising its ability to provide old age security, and (ii)continueon-goingeffortstoimproveadministrativeefficiency. 88. Strengthen assistance to the unemployed through the following actions: (i)re-design the unemployment benefit program to restrict it to laid off workers, phasing out benefits to new labor market entrants or re-entrants. Consideration may be given to replacing the insurance program with a flat benefits program funded by general tax revenues. Such a program would be simpler to administer, and progressive. Additionally, by depending on general revenues it would help reduce payroll taxes. Adequacy and predictability o f benefits should be ensured. (ii)conduct impact evaluations of existing active labor market policies, and phase out the less cost-effective ones (such as job creation programs which are micro credits provided through the banking system at beneficial terms). Although they are more expensive, limited use of self targeted public works programs could provide an effective safety net in some rural areas or poorer regions. 89. Further efforts could improve the targeting of social assistance benefits to the poorest. The Government plans to phase out social privileges from 200323,which would free resources for better targeted programs. Further efforts should be made to reduce errors o f inclusion and exclusion o f all poverty benefits programs, while serious consideration should be given to consolidation of programs. Conditional cash transfers (for example to encourage school attendance or use o f preventative health care services by the poor) could also be considered. 90. Reduce the social costs of reforms through an effective safety net. These reforms include restructuringandthe associatedemployment loss, or the price liberalizationreforms noted above. To reduce social costs of layofls, labor re-deployment programs (which typically combine passive programs such as severance pay, with active programs) can help alleviate the social costs of restructuring in particular industries or as a general policy. Existing unemployment benefit, effective active labor market programs, and social assistance programs could also be used to protect these workers and their families against low incomes as a result o f layoffs and redundancies. Toprotect thepoorfFom cost of living increases, use means tested social assistancethrough mahallahs. To allay the costs of utilities price increasesfor thepoor, consider various policy options. For example, life-line tariffs that provide subsidy to the poor, while recouping costs from higher income groups, could be 23A resolutionlargely abolishingsocial privileges was passed inApril 2003 29 used where there is universal metering(electricity). For nonmetered utilities (such as gas), the poor could be protected through the use o f existing social assistance programs or notional burden programs (which compensate for a portion of utility services exceeding some proportion of family income), or one off transfers deliveredthrough the mahallahs. 4. Building inclusiveinstitutionsand facilitatingempowerment 91. Social and economic policies can go a long way to create opportunities, reduce poverty and bring about a more egalitarian society. But in order for these policies to work, institutions must work well and in the interests of all parts o fthe society, including the poor. There are three main areas for improvement. 92. Increase the accountability of state institutions and reduce corruption. The high incidence of informal payments or bribes to public officials constitutes a disproportionate burden for the poor, and is indicative of weak accountability o f state institutions. Continued reduced burden of regulations, tax reduction and elimination of discretion inthe tax regime, will all reduce the incentives for seeking or making informal payments. Improving clarity o f regulations and policies i s another important action. Improved civil service pay, and the introduction o f performance based salary structures, strengthened oversight and monitoring, could further reduce incentives for informal payments to enhance low pay. 93. Increase transparency, access to information and dissemination of policies and rights. This is a key element o f any strategy to enhance accountability o f state institutions. Improving the quality o f regulation and eliminating lower level regulations subject to greater discretion will improve transparency, as will efforts to increasedissemination ofpolicies with the public. Inaddition, improvingthe legal framework for improved access to information, and for a free media will be important to enhance oversight by the public. 94. Undertake measures to enhance voice andparticipation and empower communities, especially the poor, such as encouragement of civil society organizations, and representatives of consumers and businesses. Participation could be enhanced, and the relevance of policy improved by increasing public consultation, discussion and debates (for example on informal payments), directly with communities or with their representatives. The Government has, in recent years increased the role of communities in decision making, for example in delivering social assistance through the mahallah committees. The role o f local mahallahs or other NGOs could also be enhanced to represent rural communities in mediation or in policy formulation. However, given the role that mahallahs themselves play in the delivery o f social assistance benefits, additional measuresto enhancetheir accountability may need to be introduced. 5. Improving Monitoring of Living Standards 95. Monitoring living standards will help policy makers review impact o f macro economic developments and economic and social reforms on household welfare, and to revise such policies accordingly. The main process involves improving capacity for collection and analysis o f household and administrative data, and strengthening the linkage between data analysis and policy formulation. It is equally important to fully disseminate individual record survey data (while withholding exact name and address o f respondents to promote confidentiality and true reporting). The key survey instruments for poverty monitoring include: 0 TheFamily Budget Survey.The recently improved Family Budget Survey has made it possible to provide the first national level assessment o f poverty in Uzbekistan, and the recent adoption of the improved survey for regular collection of data is an important accomplishment for the poverty monitoring system in Uzbekistan. It has demonstrated the additional value of survey based information on income and non-income dimensions of poverty. The utilization of this data source for poverty analysis, however, has provided important feedback on how this survey might be improved. Key measures include, strengthening o f the survey instrument to measure consumption based poverty (to more accurately capture non-food expenditures 30 among households), and expanding the present relatively limited information on non-income dimensions of poverty. Labor Force Surveys. Regular labor force surveys would provide regular updates, based on international standards, on the employment and unemployment developments inthe country. As a first step, the Ministryof Labor and Social Protection introduced a quarterly labor market monitoring system in2002. The Population Census. This is overdue in Uzbekistan with the last census having been conducted in 1989. An updated population census is essential for improving the reliability of the sampling frame for the household surveys Poverty Mapping. An updated census in combination with the Family Budget Survey could be used to potentially carry out cost-effective poverty maps like those that are currently being used in many other transition countries. Such poverty maps could be of great value in improving sub-regional targeting o f programs. 31 41 0 m m d m v1 sE rc s, 0 Y c 0 v1 in Q 0 P B .- e L