Report No. 29619-TU Turkey Poverty Policy Recommendations (In Two Volumes) Volume II August 8, 2005 Human Development Sector Unit Europe and Central Asia Region Document of the World Bank Table of Contents ACKNOWLEDGEMENT EXECUTIVESUMMARYOF VOLUMETWO 1. Brief Summary o fVolume One........................................................................................................................... 1 A . Data Comparability.............................................................................................................................................. i B. Recent Economic Developments ........................................................................................................................ ii c. PovertyProfile.................................................................................................................................................... ii 11. Key Findings...................................................................................................................................................... iv Policy Recommendations ................................................................................................................................... v A. 111. Macroeconomic .................................................................................................................................................. v B. Health.................................................................................................................................................................. v C. Education............................................................................................................................................................ v D. vi E. Labor.................................................................................................................................................................. Social Protection................................................................................................................................................ v i PARTI:VOLUME ONE SUMMARIZED................................................................................................ 1 A . 1 B. Data Comparability ............................................................................................................................................. Recent Economic Developments........................................................................................................................ 2 c. Poverty Profile.................................................................................................................................................... 3 PART11: KEYFINDINGS........................................................................................................................ 13 A. Turkey is a Medium- to High-Inequality Country ............................................................................................ 14 B. Extreme Poverty inTurkey is Low................................................................................................................... 14 c. Poverty i s High Compared to EuropeanUnion Countries ................................................................................ 14 D. Regional Differences are substantial ................................................................................................................ 15 E. Inequality acts as a "Brake" on Poverty Reduction .......................................................................................... 15 F. Poverty Projections in2004.............................................................................................................................. 16 PART1II:POLICY RECOMMENDATIONS............................................................................................ 20 G. Macroeconomic Indicators and Poverty ........................................................................................................... 17 A. Macroeconomic Policy Recommendations and Government ReformProgram ............................................... 20 B. 23 C. HealthPolicy Recommendations ...................................................................................................................... Education Policy Recommendations ................................................................................................................ 26 D. 27 28 References ................................................................................................................................................................... E. Labor Market Policy Recommendations........................................................................................................... Social Protection Policy Recommendations ..................................................................................................... 32 List of Tables Table 1. ProjectedIncidence o f Poverty with Different GrowtWInequality Scenarios .............................................. 15 Table 2. Projected Poverty Gap with Different GrowtWInequality Scenarios............................................................ 16 Table 3. Turkey: Poverty Projections ........................................................................................................................ 17 Box 1. Tax Policy ....................................................................................................................................................... 22 List ofAnnex Tables Table 1. Inequality as Measured by the Gini Coefficient; ......................................................................... 33 Table 2 Per Capita GDP and Poverty . ....................................................................................................... 34 This volume is a product o f the staff of the International Bank for Reconstruction and DevelopmenVThe World Bank . The findings. interpretations. and conclusions expressed inthis paper do not necessarily reflect the views o f the Executive Directors o f the World Bank. or the governments they represent. The World Bank does not guarantee the accuracy of the data included inthis work. The boundaries. colors. denominations. and other informationshown o n any map inthis paper do not imply anyjudgment o n the part o f the State Institute o f Statistics and the World Bank concerning the legal status of any territory or the endorsement or acceptance o f suchboundaries. ACKNOWLEDGEMENTS The first volume i s ajoint report o f the World Bank and State Institute of Statistics (DIE). This second volume i s a World Bank report. Volume One o f the Joint Poverty Assessment Report (JPAR) represents the basic findings incommon between the State Institute o f Statistics and the World Bank, drawing mostly from the 2002 Household Budget Survey (HBS). Volume Two is of World Bank authorship only, and represents the views o f the World Bank in considering Volume One and findings from previous World Bank and Turkish literature. DIE does not endorse or have any responsibility for the policy recommendationshereinVolume Two. Volume Two: Managed byJeanine Braithwaite (ECSHD) with contributions o fthe World Bank team members from Volume One and the ECSPE team working in Turkey. Edited by Diane Stamm (Consultant). Processed by Carmen Laurente, Jennifer Manghinang, and Gizella Dim (ECSHD). The team benefited from comments of peer reviewers Jeni Klugman (AFTP2), Peter Lanjouw (DECRG), and Edmundo Murmgarra(ECSHD). EXECUTIVESUMMARYOFVOLUME TWO The TurkeyJoint Poverty Assessment Report (JPAR) consists of two volumes. The first volume, titled Turkey Joint Poverty Assessment Report, is a joint publication of the State Institute o f Statistics (DIE) and the World Bank. The State Institute o f Statistics (DIE) o f Turkey and the World Bank have issued ajoint report on poverty in Turkey for the first time. The findings o f Volume One have been subject to thorough technical review by both institutions, and are considered to bejoint findings. This second volume is entitled "Turkey: Poverty Policy Recommendations," and is a World Bank report. DIE does not endorse or have any responsibilityfor the policy recommendations in Volume Two. Additionally, the conclusions inVolume Two that are based on Volume One are attributed only to the World Bank. Volume Two consists of three sections: (i) an Executive summary o f Volume One; (ii) key findings of Volume One (World Bank conclusions from the analysis o f Volume One); and (iii)World Bank policy recommendations in the sectors considered inVolume One (macroeconomy, education, health, labor andsocial protection). I. BRIEFSUMMARYOFVOLUMEONE Volume One sets out a new poverty line' methodology for Turkey as the basic measure of poverty in the country. However, several poverty lines are calculated for the purpose o f international comparability, and comparability to the World Bank's poverty measures using the 1987 and 1994 data. In2002, 27 percent o f the Turkish populationwas poor, based on the new poverty line methodology detailed in Annex One o f Volume One (food and non-food consumption). However, very few (nearly zero) consumed under the food line or under the $1 perpersonperday lineused ininternational comparisons (Table A.I.8 found inVol. 1). A. DATACOMPARABILITY An in-depth analysis of the 2002 Household Budget Survey (HBS) compared to that from 1994 shows that living standards in Turkey remained almost unchanged. Poverty based on the previous methodology declined gradually from 1987 to 2002, from 38.5 percent to 34.5 percent. Poverty based on the updated methodolog9 declined from 28.3 percent to 27 percent between 1994 and 2002. On the other hand, inequality marginally increased. Extreme poverty, already low, further declined from 1994 to 2002. Food poverty declined from 2.9 to 1.4 percent, while $1 per person per day poverty, depending on purchasing power parity (PPP) used, was 2-3 percent or evennegligible (0.2 percent). ` Thepoverty line i s the minimumamount o f consumptionneeded for an individual or household to cover its basic needs for food and non-food goods. This is the updated poverty line for 1994, butthe consumption aggregate for 1994 is the old version. For 2002, both the consumption aggregate and the poverty line were the new methodology. Additional details are found in Chapter One (Data Comparisons) and Annex One (Methodology) o f Volume One. B. RECENT ECONOMIC DEVELOPMENTS Despite many advantages ranging fi-om its strategic location to its dynamic population, Turkey has not achieved the stable highgrowth o f leadingemerging market economies. On average, the Turkish economy grew slightly under 3 percent per year over the past decade, well below the best performing emerging economies. Macroeconomic instability, rooted in fiscal imbalances, has played an important role inTurkey's inabilityto grow as fast as other countries. Inresponse to the crisis following the collapse of the crawling peg and subsequent devaluation, the Government announced a strengthened economic program in M a y 2001. Economic activity rebounded strongly in 2002, and the recovery continued into 2003, with real GNP growth reaching 7.9 percent in 2002 and 5.9 percent in 2003. While the inventory build-up led the recovery in 2002, private consumption and investment was behind the strong growth performance in 2003. Fiscal gains were significant in 2003, as the primary surplus rose from 4 percent o f GNP in2002 to over 6 percent o f GNP in2003. C. POVERTYPROFILE Poverty in Turkey is strongly associated with age and household composition; children and families with children are poorer than average. Poverty rates are higher for unemployed heads, and female heads, and sharply higher for illiterate heads. There is a sharp difference in the poverty rates between rural and urban households, with rural poverty substantially greater. Poverty restricts the poor from accessing many goods and services. The poor are less able to afford discretionary expenditures. Turkey is a middle-income country, and its inequality is high. Both consumption and income indexes indicate that inequality is higher in urban areas than in rural areas, but not much. Other data confirm overwhelmingly that there is a sharp East-West divide in Turkey where Southeastern and Eastern Anatolia regions are much poorer and have sharply lower human development indicators than Western Turkey. Education Inthe 1997-1998 school year, Turkey shifted from five years o f compulsory schooling to eight. Enrollment rates increased soon after this reform, not only for the 8-year basic education cycle but also for secondary education. As a result, poor households receive a larger share o f public expenditures on basic and secondary education. The level o f public spending on education also increased significantly after 1998, both in real terms and as a percentage o f gross domestic product (GDP). While rapid progress in recent years is encouraging, analyses o f household survey data reveal that there is significant room for further pro-poor improvements when it comes to: (i) quality o f schooling-the parents o f poor children are more likely to report problemswith schools, and they are less likely to be satisfied with the quality of education their children receive; and (ii)enrollments in secondary education-the poor, females, and children who do not have a secondary school intheir residential area are all substantially disadvantaged. * . 11 Health Turkey's health care delivery and financing system is fragmented and despite significant efforts, the service delivery network remains highly uneven, with major concentrations in urban areas. Both Household Consumption and Income Survey (HCIS) and Household Budget Survey (HBS) data suggest that over one-third (36 to 37 percent) o f the population in Turkey does not have access to health insurance, and almost half the population in rural areas remains without any coverage. Despite considerable progress achieved in the recent past, Turkey continues to rank far behind most middle-income and European Union (EU) accession countries on key health indicators. Health outcomes vary significantly across regions, reflecting the uneven supply o f and access to health care in various parts. Low-income groups suffer from significant access problems. The prime reason for not seeking care when sick, or not seeking hospital admission when required, was lack o f affordability. Public expenditure on health care grew at an average annual rate o f 7.3 percent between 1999 and 2003. Public sector spending on health care is skewed in favor o f the upper-income groups, particularly spending on outpatient care. The poor continue to face significant access barriers to health. Labor InTurkey, as inmost countries, poverty is closely related to employment status andthe type o f job, whereby informally employed and casual workers have a higher rate o f poverty. Education plays a key role in explaining employment and poverty outcomes. Unemployment in Turkey was 10.3 percent in 2002. There are very sharp differences in labor market participation rates between men and women, with extremely low rates o f female labor market participation in Turkey. Reasons for not seeking a job varied from factors relating to age or family structure (student, housewife, elderly) to disability or seasonal employment. Poverty rates o f those who had permanent employment were lower compared to those with casual or temporary jobs. Poverty was found to be sharply associated with the lack of registration at a social security institution. The largest sector o f employment in Turkey is agriculture, which is also the sector with the highest poverty rate ofthose employed init. Social Protection Social protection in Turkey consists primarily o f limited formal systems in pensions and social assistance, supplemented greatly by informal mechanisms. Formal elements o f social protection are the pension (social security) system, and the Social Assistance and Solidarity Encouragement Fund(SYDTF) and its 931 affiliated Social Solidarity Foundations (SYDVs). Institutional, care i s provided by Social Services and Child Protection Organization (SHCEK). Turkey's social security system is highly fragmented. The three pension programs are SSK (Social Insurance Organization), Bag-Kur (BK), and Emekli Sandigi (ES), covering workers, the self-employed, and civil servants respectively. Turkey also provides a small noncontributory benefit to those over age 65 who eam below the level o f the benefit. The pension system is showing large fiscal losses each year and is in need o f transfers from the government to cover those losses expected to be around 4 percent o f GDP in2004. *.. 111 The SYDTF is an extra budgetary fund financed by earmarked taxes and administered by a Cabinet Minister. The SYDTF, together with its local affiliates, is the largest programo f social assistance inTurkey interms of number of beneficiaries. Additional social protectionprograms are supported by a World Bank loan: conditional cash transfers and local initiatives. Conditional Cash Transfers (CCTs) are a national program recently introduced in Turkey with World Bank support. CCTs are payments made to the mothers o f poor children, provided they attend school or visit healthclinics. Supported under the Social RiskMitigationProject (SRMP), the Government o f Turkey has undertaken a significant expansion o f the microprojects traditionally carried out by the SYDVs with approval from the SYDTF, along with a tightening of procedures. Additionally, institutional development activities to strengthen the SYDTF, SHCEK, and DIE (especially on poverty monitoring and measurement) have been undertaken in the institutional development component o fthe SRMP. 11. KEY FINDINGS The following key findings emergedfrom the 2002 HBS data: 0 Turkey i s a medium-to high-inequalitycountry. 0 The rate o f extreme povertyinTurkey i s quite low. 0 However, the overall rate o f poverty is high when compared to the EU, and this reflects the first fact on inequality. 0 Regional differences are the major driver of country inequality, and poverty i s highly concentrated inEastemand SoutheastemAnatolia. 0 High and persistent inequality has constrained Turkey's ability to benefit from growth years, while macroeconomic instability has hindered Turkey from escaping from its medium- to -high-inequality, relatively highoverall, poverty situation. Inequality. Turkey's Ginifor income for 2002 was 44, which i s noticeably lower than the very highGinisinmanyextremely low- incomecountries. However, Turkey doesnotmeasureup well against WesternEuropewhich is substantially less unequal. Extreme poverty. Turkey compares very favorably with low- and medium-income countries in terms of its rate o f extreme poverty. Turkey has essentially a zero povertyrate for the PPP US$l perpersonper day poverty line. Overall poverty. Turkey's poverty rate of 27 percent is in the midrange for most medium- income countries but i s muchhigher than those inthe EuropeanUnion countries. Regional differences. Regional differences are the major driver o f inequality in Turkey, and by every source of information, quantitative and qualitative, Eastem and Southeastem Anatolia are markedlypoorer thanWestern Turkey. Inequality/growth. As demonstrated by the growth-inequality decomposition for 1994 to 2002, inequality has held back improvements inpoverty that growth would have led to in the absence of inequality. Projections for this in the future demonstrate that growth without redistribution iv would not "solve" Turkey's poverty problem. While sustained growth is essential for poverty reduction, if inequality is not held back, gains from growth can fail to significantly reduce poverty. And ifinequality worsens, then gains from growth can be partly or even totally erased. Several scenarios for poverty in 2004 were calculated, depending on inequality not worsening from 2002-2004, and these demonstrate that gains from growth could be considerable in the short runifhouseholdconsumption responds quickly or fully to GDP growth. 111. POLICY RECOMMENDATIONS A. MACROECONOMIC The JPAR and the previous poverty assessment emphasize that the creation of well-paying formal sector jobs is the main lever for poverty reduction in Turkey, and that macroeconomic instability has severely limited Turkey's attempts to reduce poverty since 1987. The Turkish Government has undertaken a variety o f policy reforms that are intended to promote stable economic growth which are being supported by the World Bank. These reforms will create the conditions necessary for sustained economic growth. Additionally, the Government's reform program includes measures on strengthening the tax system, including the elimination of loopholes and increased transparency, equity, andprogressivity o f taxes. The tax strategy should assist in addressing Turkey's persistent income inequality, which has been a brake on poverty reduction. The Government has adopted a strategy for tax policy reform that would bring Turkey's tax regime in line with best practice in the European Union. Implementation o f the Government's tax reform strategy is crucial for many reasons, not the least o f which would be to address Turkey's high level o f income inequality. Decomposition and simulations demonstrated that by addressing inequality, the poverty reduction effects o f growth could be substantial. A transparent and progressive tax regime would enable Turkey to achieve other economic reform goals, such as expenditure targets for the social sectors, which in turn would help address the access issues documented inthe health and education chapters o f thejoint volume. B. HEALTH Reforms in four areas could substantially help improve lower-income group access to health care, while contributing to an improvement o f the country's health outcomes. These include: (a) expansion o f health insurance to assure better coverage o f low-income households; (b) undertaking steps to assure that disadvantaged areas with poor health outcomes have access to professional health care services; (c) strengthening the quality o f primary health care services, particularly in underserved areas; and (d) strengthening quality and outreach o f maternal and child health services and essential preventive care. c. EDUCATION In order to ensure long-term economic growth and equity, there is a need to maintain public education spending levels o f at least 4.25 percent o f GDP (which is in fact a target stipulated in various Ministry o fNational Education publications). The gross enrollment ratio for compulsory 8-year basic schooling is 97 percent. Enrollment rates could increase even further if children of parents with little or no schooling (the demographic group most likely to be out o f school) were V encouraged to enroll in school. If compulsory schooling were further increased to 12 years, as mentioned in some Government planning documents, the beneficiaries would be almost exclusively from the most vulnerable groups. Nonetheless, the prerequisites to such a move include raising the efficiency in the utilization of existing secondary schools and perhaps, establishing new secondary schools. The quality o f education also needs improvement especially for the poor: parents o f poor children are more likely to report problems with schools, and less likely to be satisfied with the quality o f education their children receive. D. LABOR Implementing the reforms o f the education system and the conditional cash transfers o f the Social Risk Mitigation Project (SRMP) should help more educated women to participate in the labor market in the future. Also, the literacy training and micro-project endeavors under the Local Initiatives component o f the SRMP should assist in this long-term goal. On the linkage among informal and agricultural employment and poverty, here the best policy recommendation is for Turkey to stay the course on macroeconomic reforms that have already led to significant real economic growth. As the economy grows and develops, more formal andbetter-paying jobs will emerge from that growth, andthese will provide the avenue for Turkey's 27 percent poor to find betterjobs that will enable them to escapepoverty. E. SOCIALPROTECTION Recommendations for reform o f pensions include: a rise inthe retirement age, both in the short and longrun; a rise inthe number of years o f contribution required before receiving a benefit; a reduction inthe benefit accrual rate is also required to bring Turkey in line with other countries; a move to wage indexing the pensionable salary in lieu o f the current valorization by nominal GDP growth; maintaining price indexation o f pensions as dictated by the 1999 Social Security Law, in order to help contain costs; eliminating and avoiding supplementary payments to pensioners; maintaining or reducing the contribution rate; unification o f the pension schemes, both in terms o f benefit structure and in terms o f administration to improve equity and administrative efficiency; and integrationof social pensions with other social assistance benefits. Recommendations for reform o f social assistance include: restructuring the system in one single agency/ministry responsible for determining eligibility and payment o f cash benefits (namely the SYDTF) and decentralizing service provisions to make them more responsive to local demands; financing the system adequately; improving the targeting o f social assistance; and expanding CCTs. vi PART I: VOLUME ONE SUMMARIZED The basic data used inVolume One are from the official 2002 Household Budget Survey (HBS), conducted by DIE. Comparisons over time are made to the 1987 and 1994 official DIE HBS surveys. Additional qualitative information was gathered from a variety o f primary and secondary sources. Limited quantitative data from an unofficial survey o f 2001 are used as secondary sources in some o f the chapters. Administrativedata from other Turkish Government agencies are also used. The 2002 HBS sample was designed to be representative o f the population of Turkey, and to provide reliable information needed for an urban-rural breakdown o f data. It was not designed to be regionally representative. Thus, only qualitative data and secondary source data are used herein to discuss the regional dimensions o f poverty in Turkey. The World Bank and DIE anticipate a further joint report on regional aspects of poverty inTurkey, based on data from the 2003 HBS, which has a larger andregionally representative sample. This report sets out a new poverty line3methodology for Turkey as the basic measure o fpoverty in the country. However, several poverty lines are calculated for the purpose o f international comparability, and comparability to the World Bank's poverty measures using the 1987 and 1994 data. The basic findings for Turkey for 2002 were published in a press release by DIE (April 13, 2004), and this report provides the underlying analysis and methodology for these figures. In 2002, 27 percent of the Turkish population was poor, based on the new poverty line methodology detailed in Annex One (food and non-food consumption). However, very few (nearly zero) consumed under the food line or under the $1 per person per day line used in international comparisons (Table A.I.8 found inVolume One). The analysis inthis report refers generally to the new poverty line methodology that results in27 percent poor. This line is called "complete" poverty line, and is referred to as "Total poverty" in statistical tables. Additional poverty lines and rates can be found inAnnex One o f Volume One. A. DATACOMPARABILITY An in-depth analysis of the 2002 Household Budget Survey (HBS) compared to that from 1994 shows that living standards in Turkey remained almost unchanged. Poverty based on the previous methodology declined gradually from 1987 to 2002, from 38.5 percent to 34.5 percent. Poverty based on the updated methodology" declined from 28.3 percent to 27 percent from 1994 to 2002. On the other hand, inequality marginally increased. Extreme poverty, already low, further declined from 1994 to 2002. Foodpoverty declined from 2.9 to 1.4 percent, while $1 per The poverty line is the minimumamount of consumptionneededfor an individualor householdto cover its basic needs for food and non-food goods. This is the updatedpoverty line for 1994, butthe consumption aggregate for 1994is the old version. For 2002, both the consumptionaggregateandthe poverty line were the new methodology. Additional details are found in Chapter One (Data Comparisons)and Annex One (Methodology) of Volume One. 1 person per day poverty, depending on purchasing power parity (PPP) used, was 2-3 percent or even negligible (0.2 percent). A poverty-growth decomposition demonstrated that while economic growth was a main driving force in poverty reduction, much o f the gains from growth were offset by inequality, which slightly worsened from 1994 to 2002. These conclusions should be treated with caution in that both 1994 and 2002 were either crisis years, or immediately following crises, and so there was no measurement o fthe effect o f sustained growth on poverty. DIEis now conducting surveys to measure poverty annually, so the measurement problem should abate inthe future. B. RECENT ECONOMICDEVELOPMENTS Despite many advantages ranging from its strategic location to its dynamic population, Turkey has not achieved the high growth o f leading emerging market economies. On average, the Turkish economy grew slightly under 3 percent per year over the past decade, well below the best performingemerging economies. Turkey has suffered from an exceptional degree of macroeconomic instability characterized by highinflation andsharp swings inbusiness cycles. Inflationwas higher and growth lower inthe 1990s than in the 1980s. Inthis period, unsustainable fiscal policy has repeatedly put pressure on the Turkish Lira (TL), fueled inflation, and undermined financial stability. Open capital accounts and a poorly regulated banking system have magnified the impact o f unsustainable fiscal policy on macroeconomic stability. Short-term capital flows have fluctuated widely as investors responded to the boom-bust cycle driven by unstable conditions. In2000, a crawling pegexchange rate regimewas launched to ridthe economy ofinflation. Key structural reforms in social security, infrastructure, agriculture, privatization, and banking were introduced. However, these achievements were insufficient to avoid a crisis, given the extent of Turkey's underlying fiscal and financial sector weaknesses built up over decades of instability anddelayed reform. By 2001, persistent doubts about the peg, and underlying fiscal instability, led to a full-blown speculative attack against the currency. Interest rates shot up to several thousand percent, forcing the Government to abandon the crawling peg and float the TL. The TL immediately lost 40 percent of its value ina single day. Inresponse to the crisis, following the collapse o fthe crawling peg and subsequent devaluation, the Government announced a strengthened economic program in M a y 2001. The key structural and social elements of the program were: (a) a macroeconomic framework designed to restore financial stability and ensure public debt sustainability; (b) a rapid restructuring of the banking sector; (c) a public sector reform program; (d) renewed privatization; and (e) enhanced social assistance. The Turkish economy started to grow at a fast pace in 2002. Economic growth reached 5.9 percent in 2003, following 7.9 percent growth in 2002. While the inventory build-up led the recovery in 2002, private consumption and investment was behind the strong growth 2 performance in 2003. The current account deficit widened to almost 3 percent o f GNP in 2003, but was easily financed by short-term capital inflows, public sector borrowing abroad and reverse currency substitution. Inflation fell to 18.4 percent in 2003, and the latest data suggest that inmid-2004, inflation declined to the important single digit level for the first time since the 1970s. Aggregate unemployment remained stable at around 10percent but this was helpedby a temporary shrinkage in the labor force. Fiscal gains were significant in 2003, and the primary surplus rose from 4 percent o f gross national product (GNP)in2002 to over 6 percent o f GNP in 2003, close to the programmed 6.5 percent target. Monetary policy followed a policy o f implicit inflation targeting, with the Central Bank o f Turkey (CBT) occasionally intervening in the foreign exchange market to dampen what was deemed to be excessive fluctuation in the exchange rate. The decline in inflation, which was aided by the strength o f the TL, led to a commensurate decline in interest rates from a nominal 60 percent inthe first quarter o f 2003, to about 25 percent early in2004. However, deterioration in the global financial indicators in early M a y 2004 combined with the higher than expected current account deficit figures have led to a sharp weakening in domestic financial indicators. The excess volatility in the foreign exchange market was curbed to some extent by the Central Bank's intervention. It appears that some relative stability has been achieved in domestic financial markets. These recent deteriorations have underlined Turkey's exposure to shocks from the external environment. While depreciation o f the TL is an adjusting factor to the deteriorating current account balance, it is also likely to affect inflationary expectations and domestic interest rates. Higher domestic interest rates, in turn, together with the impact o f the TL's depreciation, would influence the overall fiscal deficit and economic activity. The spillover from global liquidity tightening and rising spreads are likely to increase the cost o f external borrowing. Such developments, ifpersistent, could disrupt the virtuous cycle that the economy has experienced over the last year and a half. C. POVERTYPROFILE Household Size and Composition. Poverty in Turkey is strongly associated with age and household composition; children and families with children are poorer than average. Poverty increases monotonically with additional household members, starting at three members. Larger households are poorer than smaller households, and this is primarily due to the fact that the additional household members are more likely to be children, who have a higher poverty rate. Households with no children or only one child had poverty rates below the average. There is also a correlation between having elderly members and household poverty, though this correlation i s not as marked as with having additional children. Having one more elderly member did not appreciably increase the risk o f poverty, and having two or more slightly elevated the risk. With respect to correlation with age, younger children are poorer, active-aged adults are not as poor on average, and the elderly are poorer than adults, but not as poor as children. Household Head Characteristics. The household head has a substantial impact on the poverty status of his or her household, through the employmenthactivity nexus and the amount of income she or he can contribute to the household. The poverty rate for households with 3 unemployed heads is 35.4 percent, compared to 27 percent overall poverty rate. Besides this, other demographic features that are associated with poverty include whether the household head is a female or not (32 percent poor versus 26.6 percent poor for female and male heads respectively), and the education of the household head. Illiterate heads and those who had not completed primary school heads hadpoverty rates nearly twice the average, while those few with masters or other advanced degrees had a poverty rate o f zero. Spatial Characteristics. There is a sharp difference inthe poverty rates between rural and urban households; the poverty rate is nearly 35 percent for the rural population, but only 22 percent for the urban population. The factors for highrural poverty are the same as for poverty overall. Rural areas are characterized by limited employment opportunities, and rural households where the head is unemployed face a substantial risk o f poverty. Other kinds o f inactivity have different implications on rural and urban areas. The major driver for rural and urban employment findings appears to be sector o f employment, where rural location is dominated by agriculture, which offers less lucrative options compared to formal employment found in urban areas. Education has identical effects on both rural and urban areas, whereby those who are illiterate or whose education is limited to primary school have higher poverty rates than average, and graduates o f higher education are much less likely to be poor. In both areas, poverty rates steadily decrease as years o f education increase. Non-Income Aspects of Poverty. Poverty restricts the poor from accessing many goods and services. Non-income here refers to material items, assets, or services that are ultimately obtained through income. InTurkey, there is a marked difference inthe kindof dwelling. About half the sample population lives in a house, and another 27 percent lives in apartments. While individual houses are primarily in rural areas, apartments are almost exclusively an urban phenomenon. Only 6.5 percent o f the apartment dwellers are poor, while 36 percent of those who live inhouses are poor. Interms of landownership, 27 percent of the sample population reported they had a field, but these households were poorer than average. The mean size of the fields for poor households was 75 percent o f the non-poor fields. More than one-fifth o f the population possessed a car, but only 6 percent o fthese were poor. The poor are less able to afford discretionary expenditures. The poverty rate o f those who smoke, drink, take computer courses, and have access to public transportation to school, for example, is lower for all than the overall poverty rate o f 27 percent poor. Shopping pattems also vary between poor and non-poor; the latter are more likely to shop at markets or bazaars. Inequality and Regional Dfferences. Turkey is a middle-income country, and its inequality is high. Both consumption and income indexes indicate that inequality is higher in urban areas than inrural areas, but not much. Other data confirm overwhelminglythat there is a sharp East- West divide in Turkey where Southeastem and Eastern Anatolia regions are much poorer and have sharply lower human development indicators than Western Turkey. 4 Education In 1923, the year in which the Republic of Turkey was founded, the adult literacy rate was approximately 10 percent. Such a low starting point (not uncommon for that era) gave the Republic a major cause to introduce a series of ambitious reforms that included a move toward secular education and the adaptation o f the Latin alphabet. While the schooling environment in Turkey gradually improved over time, the 1997-1998 school year marked another major leap forward, in that, compulsory schooling increased from 5 years to 8 years for children aged 6 to 14. Enrollment rates increased soon after this reform, not only for the 8-year basic education cycle but also for secondary education. Public Spending on Education Turkey's public spending on education increased significantly after 1998, both inreal terms and as a percentage of gross domestic product (GDP). Consequently, as o f 2000, Turkey's public spendingon education as a percentage of GDP has become comparable to the spendingpattems observed incountries o f similar levels o f economic development. The expansion o f compulsory schooling to 8 years had an extremely positive impact on the distribution of public spending on education across rich and poor households. In 2001, 21.7 percent o f public spending on basic education reached the poorest 20 percent o f households (as opposed to 15.8 percent in 1994). But there is significant room for improvement when it comes to secondary education, since only 13 percent o f public spending reached the poorest 20 percent o f the population in 2001. Household expenditures on education strongly reinforce the disadvantageous situation o f poor children: in 2002, the wealthiest 20 percent of households spent 6.4 times more on education than did the poorest 20 percent of households. The current situation represents a significant improvement compared to 1994, when the wealthiest households spent 28.8 times more on education than didthe poorest households. InefJicient ResourceAllocation in the Education Sector After the expansion of compulsory schooling, the number of primary school students and students enrolled in general secondary schools has significantly increased. The number of studentsenrolled invocational secondary schools has remained roughlythe same. Classroom construction at the basic education level has been impressive: students per classroom at this level declined between the 1997-1998 school year and 2002-2003 school year despite increasing enrollments. At the secondary level, however, a curious trend is present inthat: (i) the number of vocational school classrooms increased significantly in 2002-2003 even though enrollment in vocational schools did not increase in recent years; and (ii) the number of general secondary classrooms remained stagnant, even though general secondary school enrollment increased visibly in recent years. As a result, as o f the 2002-2003 school year, average classroom size was 18 invocational secondary schools and45 ingeneral secondary schools. 5 Enrollment in Early Levels of Schooling According to the 2002 HBS data, 97 percent o f children aged 6 to 14are either enrolled inschool or have completed the basic education cycle. Parental schooling is a very good predictor o f children's enrollment level: about 70 percent o f the children who are not attending school (but who are o f school age) have at least one parent who has not completed primary school. An econometric model o f enrollment in secondary schools reveals that males are 7 to 8 percentage points more likely to continue their education beyond compulsory schooling. Secondary school availability in the residential area has a strong predictive power, boosting the probability of enrollment by 10percentage points-only 64 percent o f households reported that a secondary school is available in their residential area. Other important correlates o f secondary school enrollment are household wealth and presence o f a mother inthe household. Universities and ThePoor InTurkey, entrance to universities is primarilybased on a student's performance in a centrally administered examination. Grades in secondary school (and a specialized field in secondary school) are other factors that influence the overall score. An analysis o f the 1997 University Student Survey found that students fiom high-income families are much more likely to be enrolled in private universities and well-established institutions. Thus, (the few) students from poor households who are enrolled inuniversities do not enroll inuniversities o f the same quality as those o fwealthier students. Some o f the other findings were that private tutoring plays a key role indetermining who attends what type o f university. As the main reason for not receiving private tutoring, 57 percent of surveyed undergraduate students mentioned lack o f economic resources. Parental ViewsAbout Quality of Education Analysis o f 2001 household survey data reveals that household members are more likely to report problems with public schools (36.7 percent) compared to private schools (25.8 percent). The lack o f books and supplies emerges as the leading problem-reported as a problem for 15 percent o f children enrolled in public schools and 10 percent o f children enrolled in private schools. The next major problem is poor teaching (inboth public and private schools), reported inroughly 10percent o f cases. The urbadruraldifferences are pronounced-only 45 percent of rural responses indicated "no problem with school" as opposedto 68 percent inurban areas. The wealthier are much less likely to report problems with school. Complaints about lack of books and supplies decline rapidly with increases in household wealth. Complaints about the condition o f facilities also decline with wealth. While complaints about poor teaching remain constant across all wealth groups, complaints about lack o f teachers decline with wealth. 6 Schooling Attainment and SelectedLabor Market Outcomes Private returns to schooling are very high in Turkey. The gender gap in earnings is visible, on average, males earn 45 percent more per hour than females with similar characteristics. Schooling has a robust, positive, and large impact on earnings for both genders. Ifone estimates separate earning regressions for males and females, the impact o f schooling on earnings is found to be more pronounced for females. In other words, while females earn less on average, the variation in earnings by schooling attainment is more significant for females when compared with males. These findings, jointly with the finding that only 15 percent o f those who report non-zero wages are females, suggest that by under-investing ingirls' schooling and by operating with extremely low female labor force participation rates, Turkey foregoes a vast potential humancapital resource that can fuel the economy. Finally, contrary to common perceptions, the unemployment rate among vocational secondary graduates tends to be at about the same level as the unemployment rate among general secondary school graduates. Health TheHealth Care System Turkey's health care delivery and financing system is fragmented. Both public and private providers supply health services. The Ministry o f Health (MOH), the Social Insurance Organization (SSK), and university hospitals are the main providers. While the public health care sector primarily predominates, private sector provisions are gaining importance in western and urban parts o f the country. Despite significant efforts, the service delivery network remains highly uneven, with major concentrations in urban areas, particularly in the western part o f the country. This skewed distribution has led to significant regional differences in access to and use o f health care and, concomitantly, health outcomes. Health Insurance Several public health insurance schemes currently provide financial protection to various target groups. SSK's health insurance is the most important one, catering to those employed in the formal sector. The green card system, introduced in 1992, is intended to provide coverage to low-income groups who are not covered otherwise. The Government soon plans to shift to universal health insurance, which would operate on the principles o f solidarity and risk pooling, andprovide coverage to the entire population. Both Household Consumption and Income Survey (HCIS) and HBS data suggest that over one- third (36 to 37 percent) o fthe population inTurkey does not have access to health insurance, and almost half the population inrural areas remains without any coverage. The green card program fails to provide broad coverage to all those living in poverty. Thus an important share o f the lowest-income households remains without access to health insurance coverage even ifthey may 7 be eligible for it, and makes significant out-of-pocket payments when seeking health care. The lack o f insurance coverage leads many low-income households to forego health care, which in tumnegatively affectstheir healthoutcomes. Health Outcomes Despite considerable progress achieved in the recent past, Turkey continues to rank far behind most middle-income and European Union (EU) accession countries on key health indicators. Health outcomes vary significantly across regions, reflecting the uneven supply o f and access to health care invarious parts. The Turkish health system faces a dual challenge. Significant parts o f the population continue to be afflicted with a high burden o f disease from preventable infectious diseases, and high matemal and infant mortality rates, typical o f developing countries. At the same time, non- communicable diseases prevalent in developed countries affect a growing share o f the population. Access Toand Useof Health Care People with health insurance, including a green card, are more likely to seek health care when ill than those without insurance. The likelihood o f seeking care when illis lower between the bottom two quintiles than among the upper quintiles, with the difference being particularly marked inrural areas. Low-income groups suffer from significant access problems. Determinants of CareSeeking The prime reason for not seeking care when sick, or not seeking hospital admission when required, was lack o f affordability. One out o f five people from the lowest income quintile who sought outpatient care reported that the main problem with the care was that it was too expensive. Lack o f a facility nearby did not appear to be a prime determinant o f not seeking care when ill. The share o f the population that had to pay for outpatient treatment, drugs, and hospitalization is higher among the lowest income quintile than among the upper-income groups. Furthermore, a multivariate analysis of the determinants o f health care-seeking behavior confirms that income, insurance coverage, household size, gender o f household head, and severity o f illness are the most important determinants o f an individual seeking health care. Health Care Expenditure Out-ofpocket Payments. Households in Turkey allocate a modest share o f their total expenditure to health care in the form o f out-of-pocket payments. Income is a major deciding factor on the amount spent. The top quintile spends about twice as high a share o f total household expenditure on health care as the lowest quintile. The largest share o f out-of-pocket expenditure on health is allocated to the purchase o f drugs, with payments for outpatient consultation rankingsecond across all income levels. 8 Public Expenditure on Health Care Public expenditure on health care grew at an average annual rate o f 7.3 percent between 1999 and 2003. The main public financiers are the Central Government and the social security institutions. According to the recent National Health Accounts study estimates in Turkey, Central Government funding accounts for over one-third o f Turkey's health care expenditure, employer contributions account for less than one-fifth, and households pay over two-fifths through out-of-pocket payments and social security contributions. Public sector spending on health care is skewed infavor o f the upper income groups, particularly spending on outpatient care. Budgetary hnds are not well targeted toward assuring equitable access by the entire population. Thus, overall, the relatively important public subsidies to health care are benefiting the middle- and upper-income households, while the poor continue to face significant access barriers to health. Labor Market in Turkey InTurkey, as inmost countries, poverty is closely related to employment status and the type of job, whether formal or informal. Informally employed and casual workers have a higher rate o f poverty. Education plays a key role inexplaining employment and povertyoutcomes. Unemployment and Labor Force Participation Unemployment in Turkey was 10.3 percent in 2002. Labor market outcomes are mostly driven by low levels of labor force participation. Those who do not work drop out o f the labor force, and are thus not captured in the unemployment rate figures. The poverty rate of the non- participants in the labor force reflects strongly the situation o f children: inactive household members younger than 15 years o f age had a poverty rate o f 35 percent, but older inactive household members had a poverty rate o f only 22 percent, under the total povertyrate. There are very sharp differences in labor market participation rates between men and women, with extremely low rates o f female labor market participation inTurkey and even decreases inthe rate for females since the 1999 level of 30 percent, down to 27.9 percent in 2002. The male labor force participation rate was 72 percent in 2002, or more than twice that o f women's. Female unemployment rates have typically been slightly lower than the male unemployment rate. This i s primarily because so few women are inthe labor force Unemployment and Inactivity In2002,35 percent o fthose aged 12and above reportedthat they hadworked inapaidjob inthe month o f the survey. The poverty rate o f these 35 percent was 25 percent. Another 43 percent o f those aged 12 and above who reported that they did not work had almost the same poverty rate (24 percent poor) as the employed. 9 Interms o f unemployment, only 7.2 percent o fthose aged 12 and above reported that they were looking for a job. Householdswith employed heads had a poverty rate of 26 percent compared to 35 percent where the householdheadwas unemployed. Reasons for not seeking a job varied from factors relating to age or family structure (student, housewife, elderly) to disability or seasonal employment. Quality of Employment In Turkey, there exists a strong association between the type of employment and the poverty status o f the individual or household. Poverty rates o f those who had permanent employment were lower compared to those with casual or temporary jobs. The relative risk o f poverty for casual workers was 3.7 times greater than for the permanently employed. Poverty rates for self- employed andunpaid family workers were higher. Poverty was also found to be sharply associated with a lack o f registration at a social security institution. Of the 35 percent who reported being employed, 32 percent were enrolled in social security. The poverty risk for people with formaljobs was the lowest for those employed by the government and instate-owned enterprises. Poverty i s associated with the size o f the enterprise. People employed in larger firms were less likely to be poor compared to people from firms with 1 to 9 people. Almost 70 percent of the respondents (aged 12 and over) reported that they worked ina firm o f 1to 9 people, andthus had a higher poverty rate. In Turkey, poverty is associated more with lack o f work than with work. The mean number o f hours worked by the poor was 43.4 per week, whereas by the non-poor, it was 46.3 hours. Also, the mean duration o f employment o f the poor inthe same low-paying job was longer than ofthe non-poor. Sector of Employment The largest sector of employment inTurkey i s agriculture. Agriculture is also the sector with the highest poverty rate o f those employed in it. Of the 35 percent aged 12 and above, 40 percent are engaged in agriculture. The next-highest poverty rate is that o f construction. The poverty rate is lowest for mining and quarrying. After agriculture, the only other significant sectors in terms o f employment are manufacturing, andwholesale and retail trade. Social Protection in Turkey Social protection in Turkey consists primarily of limited formal systems in pensions and social assistance, supplemented greatly by informal mechanisms. Formal elements o f social protection are the pension (social security) system, and the Social Assistance and Solidarity Encouragement Fund(SYDTF) andits 931affiliated Social Solidarity Foundations (SYDVs). 10 TurkishPension System Turkey's social security system is highly fragmented. Benefits and contributions depend on one's occupation. Sosyal Sigortalar Kurumu(SSK) covers the bulk of the labor force, especially private sector employees and those public sector employees who do not qualify as civil servants. Civil servants are covered by Emekli Sandigi (ES), and the self-employed and farmers are covered by Bag-Kur (BK). Also, there are separate occupational schemes that cover various other groups. Overall, 42 percent (11 million people) of the labor force is contributing to one or the other scheme. On the beneficiary side, only 29 percent (1.2 millionpeople) of the population over age 65 is receiving an old-age pension. However, almost 3 million individuals below the age of 65 are receiving pensions, primarily old age pensions, with many of the recipients considerably younger than 65. As a result, the pension system is showing large fiscal losses each year and is inneed of transfers fromthe government to cover those losses expected to bearound4percent of GDP in2004. SSK is by far the largest system, and covers mostly private sector employees. Employers contribute 11 percent of wages for pension, and employees contribute 9 percent of wages. In 1999, the Social Security Law changed most of the SSK benefit parameters. Pre-reform and post-reform benefits are discussed indetail inthe social protection chapter. Bug-Kur primarily covers the self-employed and some farmers. Contribution rates are 20 percent for pensions and 20 percent for health coverage. To combat the perennial problem of evaluating income earned, Turkey uses the system o f minimum earnings steps, which are attributed to individuals regardless ofwhat they actually earn. Bag-Kur has a very low collection rate for its contribution revenue. Workers pay very little duringtheir working years, andjust prior to retirement, they pay Bag-Kur a lump-sum equivalent to the past-due contributions with interest, andthen receive their retirement. Emekli Sandigi covers civil servants, including military personnel. Financing o f Emekli Sandigi is somewhat different from the other plans in that health insurance during working years is not covered by the pension fund. Instead, it is covered directly by the line ministries in which the civil servants are employed. Another difference between Emekli Sandigi and other schemes is that the basis for contributions and the basis for benefits are different. Contributions are paid on the basis of basic salary. On the other hand, when pension benefits are paid, they are paid on full remuneration. Thus, there is both a financing gap and an equity issue, where lower-grade workers pay contributions on a larger share of their salary than higher-grade workers. Noncontributory Pension Benefits. Turkey also provides a small noncontributory benefit to those over age 65 who earn below the level ofthe benefit. 11 Social Assistance and Solidarity Encouragement Fund The Social Assistance and Solidarity Encouragement Fund (SYDTF) i s an extra budgetary fund financed by earmarked taxes and administered by a Cabinet Minister. The SYDTF, together with its local affiliates, is the largest program of social assistance inTurkey interms of number o fbeneficiaries. Conditional Cash Transfers Conditional Cash Transfers (CCTs) are a national program recently introduced in Turkey, supportedby a loan from the World Bank,the Social RiskMitigationProject (SRMP). CCTs are payments made to the mothers o f poor children, provided they attend school or visit health clinics. CCTs are important tools that are targeted to the poorest o f the poor, many o f whom are not able to afford the out-of-pocket expenses o f sending their childrento school. When the CCT program was fully operational across Turkey in late 2004, and 1.6 million children and 7,000 pregnant women benefitedfrom the program. Local Initiatives Supported under the SRMP, the Government o f Turkey has undertaken a significant expansion o f the microprojects traditionally handledby the SYDVs with approval from the SYDTF, along with a tightening o f procedures. In 2004, 250,000 people will benefit from income generation, employment, and social service opportunities under the SRMP Local Initiatives component, which seeks to provide these people with sustainable livelihoods, thereby lifting them permanentlyout ofpoverty. Institutional Development Institutional development activities to strengthen the SYDTF, SHCEK, and DIE (especially on poverty monitoring and measurement) have been undertaken in the institutional development component of the SRMP. These activities have been to computerize and set up an internet data base for the management information system (MIS) o f the SYDTF and SYDVs, a MIS for SHCEK, and additional computers for DIE. All three institutions also received staff training programs. 12 PART 11: KEY FINDINGS The following k e y findings emerged from the 2002 H B S data: 0 Turkey is a medium- to high-inequality country. 0 The rate o f extreme poverty inTurkey is quite low. 0 However, the overall rate of poverty is high when compared to the EU, and this reflects the first fact on inequality. e Regional differences are the major driver o f country inequality, and poverty is highly concentrated inEastern and Southeastem Anatolia. e High and persistent inequality has constrained Turkey's ability to benefit from growth years, while macroeconomic instability has hindered Turkey from escaping from its medium- to -high-inequality, relatively highoverall, poverty situation. Additionally, it is striking how similar the situation in Turkey in2002 was to that in 1994, when the previous H B S data were collected, and to the situation documented in the World Bank's previous Poverty Assessment (World Bank, 2000). Indeed, virtually all o f the policy recommendations emerging out of the 2002 data analysis are echoes o f previous findings from the previous Poverty Assessment (World Bank, 2000). This reflects the fact that Turkey's povertyandinequality situation was virtually unchangedfrom 1994to 2002. K e y findings from the previous Poverty Assessment (PA) which are confirmed in the JPAR include: 0 Turkey faces a serious challenge in generating employment (labor force participation rates have further declined since the 1997 data presented in the previous poverty assessment, and there is a very low rate of female labor force participation). 0 Productivity growth is key (agriculture continues to be a poor performer). 0 Income inequality inTurkey i s high. 0 Distribution of income is fairly stable (the JPAR finds it has worsened, but only slightly). 0 Absolute poverty is low, but economic vulnerability is widespread. e Poverty inTurkey is linked mainly to education and employment status. 0 Government spending needs to be better targeted to the vulnerable. The only finding from the previous poverty assessment that requires nuance is the finding that meeting the employment challenge requires faster GDP growth. While this is certainly true, the persistence of inequality in Turkey suggests that faster growth will fail to resolve the question of how that growth is distributed, and that inequality acts as a brake on raising living standards generally, even when growth is high, as has been the case for most of the non-crises years since 1994. Several scenarios for poverty in 2004 were calculated, depending on inequality not worsening from 2002-2004, and these demonstrate that gains from growth could be considerable inthe short runifhouseholdconsumptionrespondsquicklyor fully to GDPgrowth. 13 A. TURKEYISMEDIUM- TOHIGH-INEQUALITY COUNTRY A Turkey's inequality as measured by the Gini coefficient for income to be comparable to other countries (which are predominately income-based measures) puts Turkey inthe middle range of countries, as inequality i s so high in Afica and Asia. Milanovic and Yitzhaki (2001) estimated that Gini coefficients for these were 52.1 and 61.5, respectively, while Turkey's Gini for income for 2002 was 44, which is noticeably lower. However, Turkey does not measure up well against Westem Europe. Including the United States leads Milanovic and Yitzhaki (2001) to estimate the combined Gini at 39.4. The Gini for this geographical grouping would be much lower ifthe United States and Turkey were not included. Comparing Turkey's inequality to individual Westem European countries (Annex Table l), it is clear that Turkey has markedly higher inequality than the European Union countries. Furthermore, inequality in Turkey has not decreased since 1987. Decomposing the change inpoverty (incomparable measure) for Turkey between 1994 and 2002 reveals that increased inequality over the time period led to less reduction inpoverty than could have been achieved with equal growth. B. EXTREME POVERTYIN TURKEYIS LOW Turkey compares very favorably with low- and medium-income countries interms o f its rate of extreme poverty, measured either as purchasing power parity (PPP) o f U S $ l per person or by the food component of the national line. Turkey has essentially a zero poverty rate for the PPP US$1per person per day poverty line, andthe share ofthe populationthat i s below the food-only component o f the national poverty line is 1.36 percent (adult equivalent consumption). These rates are extremely low ininternational comparisons. As documented in World Bank (2003), there is a very high degree o f interhousehold transfers and "social solidarity" that keeps the extremely poor from starvationpoverty, and these transfers (often o f food but also of housing and clothing) have resulted in a very low rate o f extreme poverty inTurkey. c. POVERTYIS HIGHCOMPARED TOEUROPEAN UNION COUNTRIES Turkey's poverty rate o f 27 percent is inthe midrange for most medium-income countries (Table 2), but this reflects the significant dispersion o f poverty rates in those countries. Turkey's poverty rate is much lower than the poverty rates in many Latin American countries, but significantly higher than the poverty rate inChina, which dominates the medium-income class by reason o f population size. In particular, Turkey's poverty rate is much higher than that in EuropeanUnion countries, and inmany o f the accession candidate countries. 14 D. REGIONALDIFFERENCES SUBSTANTIAL ARE Regional differences are the major driver of inequality in Turkey, and by every source o f information, quantitative and qualitative, Eastem and Southeastern Anatolia are markedly poorer than Western Turkey. Inthis, Turkey is not unique-many countries grapple with the problem of underdeveloped regions, including Appalachia inthe United States, and Southern Italy. There has been tremendous and rapid rural-to-urban migration inTurkey in recent years, and the share of the rural population has fallen from 49 percent in 1990 to 40 percent in 2000, with much of the rural to urban migration occumng from Eastern and Southeastem Anatolia. However, family size remains very large in those regions, so that migration has not "solved" the problem of persistent poverty. E. INEQUALITYACTS AS A `LBRAKE"ONPOVERTY REDUCTION As demonstrated by the growth-inequality decomposition for 1994 to 2002, inequality has held back improvements in poverty that growth would have led to in the absence of inequality. Projections for this in the future demonstrate that growth without redistribution would not "solve" Turkey's poverty problem (Table 1). While sustained growth is essential for poverty reduction, if inequality is not held back, gains from growth can fail to significantly reduce poverty. And ifinequality worsens, then gains from growth can be partly or even totally erased. Table 1. Projected Incidence of Poverty with Different GrowtMnequality Scenarios PovertyRatein 5 Years PovertyRatein I O Years With GrowthandInequality With Growthand Inequality 4 Inequality (Annual change) 1 J Inequality(Annual change) 1 ISource: Turkey 2002 HBS data. 1 To see these results, it is easiest to first orient on the columns and rows labeled "O%," meaning no change, in the first half of Table 3 (5 years). If there is no growth in consumption and no change in inequality in five years, then poverty remains unchanged at 27 percent. However, reading along the row labeled 0 percent, even ifthere is no growth at all, if inequality declined 1 15 percent per year, poverty would fall from 27 to 24.7 percent. This is almost equal to the effect that keeping inequality equal but experiencing a growth improvement of 1 percent per year would have on poverty-24.1 percent. Still looking at the 5-year results, notice the bottom right entry. If inequality increases at the same 3 percent rate as growth, then poverty will be almost unchanged, declining only slightly to 26 percent. This illustrates that even the beneficial effects of growth can be almost completely offset by inequality. Additionally, it is clear that if inequality can be decreased even slightly, as long as there is growth, there will be significant strides inpoverty reduction. For example, a 1 percent growth in consumption coupled with a 1 percent decline in inequality would reduce poverty substantially, from 27 percent to 22.1 percent. The impact of inequality on the poverty gap (the distance between average consumption and the poverty line) is even greater than on the poverty headcount (Table 2). Here, even when growth is positive, as long as inequality increases, then the gap is greater. Table 2. ProjectedPoverty Gap with Different GrowtMnequality Scenarios PovertyGap in 5 Years . - Poverty Gap in10 Years With GrowthandInequality With GrowthandInequality Source: Turkey 2002 HBSdata. F. POVERTYPROJECTIONS IN2004 Inequality, while high in Turkey, has been quite stable since 1994, increasing very slightly from 1994to 2002. Thus, assuming that inequality will not change from 2002 to 2004 is a reasonable assumption, and allows the projection o fpoverty rates from 2002 to 2004. This is simply usinga shorter time period, but the same assumption as the center column in the upper left hand quadrant of Table 1 (0 percent change in equality). The more pressing question for the projection is not what to assume about inequality, since an assumption o f no change is 16 empirically validated for Turkey, but rather what would the growth rate of household consumptionbe, given the actual GDP growth from 2002 to 2004 to date. Here, GDP growth rates are a good guide, but are indicatory only, since typically household consumption increases with growth, but with a lag as it takes time for the positive effects o f growth to come to households, particularly those at the lower end o f the distribution. Accordingly, five different scenarios about growth and poverty reduction in 2004 are presented inTable 3. All the scenarios depend on inequality not worsening, and demonstrate that gains from growth could be considerable in the short run if household consumption responds quickly or fully to GDP growth. Table 3. Turkey: Poverty Projections Percent Poverty Change of Rate Reduction Poverty rate in2002 (annual, percent o fpopulation) 27.0 Poverty rate in2004 Scenarios Conservative: Household consumption (HHcons) grows only 3 percent intwo years 25.3 6.3 Medium: HHcons grows at 3 and 2.5 percent respectively intwo years 23.9 11.5 High: HHcons grows at same rates as GDP, 5.9 and 5.0 respectively 21.3 21.1 Higher: HHcons grows at 5.9 in2003 and 6.0 in2004 respectively 20.9 22.6 Lagged model: Ifgrowth inHHcons follows one-year lagged GDP growth-- HHcons grows at 7.9 in2003 and 5.9 in2004. 20.2 25.2 G. MACROECONOMIC INDICATORS AND POKERTY The Turkish Government has undertaken a variety o fpolicy reforms that are intended to promote stable economic growth which are being supported by the World Bank in the Programmatic Financial and Public Sector Adjustment Loan I11(PFPSAL 111). These reforms will create the conditions necessary for sustained economic growth. However, achieving high growth rates is not a sufficient condition for reducing p ~ v e r t y .Understandingthe relationship betweenpoverty ~ and macroeconomic indicators and the transmission mechanisms involved is key in designing effective poverty reductionpolicies. Growth-Unemployment-PovertyRelationship Under certain conditions, high growth rates can be influential in reducing poverty rates. Understanding growth dynamics plays a vital role in assessing the poverty reducing impact o f any growth process. While the Turkish economy recorded high growth rates in 2002 and 2003, the unemployment rate increased from 8.4 percent in2001 to 10.3 percent in2002 and hrther to 10.5 percent in 2003, mainly due to lack of adequate job creation. Firms had generally low Neither i s it a necessarycondition, since an improvement inthe income inequality-a redistributiono f income toward the poor-will result ina reduction o fpoverty even inthe absence o f economic growth (see Table 3). 17 capacity utilization ratios in Turkey and during the recession in 2001, the annual capacity utilization ratio fell to 70.9 percent, which was the lowest ratio inthe last 15 years. In2002 and 2003, firms used their idle capacity. The capacity utilization ratio increased to 75.4 percent in 2002 and to 78.4 percent in 2003. More recently, in the first quarter o f 2004, the capacity utilization ratio reached a historical high of 78.5 percent. Given this high level and the strong demand inthe economy, there i s a clear need for firms to expand their capacities, which is likely to entail creation o f new jobs. Although production increases that end up inincreased gross fixed capital formation or increased inventories both imply positive growth rates, given that firms are likely to use their existing capacity in increasing their inventories, job creation is possible only through new investments. The strength o f the link between new investments and job creation depends crucially on the nature o f technology employed in the new production process. In an attempt to minimize the labor cost, firms might prefer more capital-intensive technologies, in which case, labor productivity will increase and job creation will not follow the increases in production. Consequently, without new job opportunities, unemployment will continue, especially if there is also a structural component o f unemployment. Empirically, in Turkey, unemployment o f the householdhead is particularly associated with poverty. In2001, duringthe recession, Turkey's GNP contractedby9.5 percent. The major factor behind the contraction was the plunge in private fixed investments by 35 percent. However, in the recovery year 2002, the main source o f growth was the increase in inventory levels. Real GNP inTurkey grew by 7.9 percent in2002 and 92 percent of the growth was due to an increase in inventory levels. This is the main factor behind the lack o f a commensurate increase in employment; while production increased, the firms stocked up inventories rather than increasing sales in an environment o f sluggish private consumption. With limited job opportunities, the poor and the vulnerable did not have any opportunity to move out ofpoverty. In2003, while an increase in inventories continued to be a key contributor to growth, gross fixed capital formation recovered by 10 percent and private consumption by 6.6 percent, promising a more positive impact on the availability o fnewjobs compared to 2002. Another factor that is important in understanding the unemployment-poverty relationship is the large informal sector in Turkey, given that the informally employed or casual workers have a noticeably higher rate of poverty. After the 2001 crisis, many employees lost their jobs and some o f those unemployed shifted to the informal sector with lower wages. In addition, some firms prefer to employ workers informally in order to avoid paying high social security premiums. Increasing unemployment ratios and slightly decreasing poverty rates, despite the high growth rates observed in2002 and 2003, may be explained partially by the existence o f a large informal sector. A pre-condition for growth to have a poverty-reducing impact is equal distribution o f the gains from highgrowth rates among the different income groups in society. Ifonly income o f the non- poor people increases with positive growth, income inequality will increase. In Turkey, on average, a 3 percent growth rate was sustained between the years 1994 and 2002, but the poverty-growth decomposition demonstrated that real growth between 1994 and 2002 was significantly offset by a slight increase in inequality, so that there was only a slight reduction in 18 poverty. During this period, non-poor people benefited from the advantages o f the positive growth more than the poor people. Additionally, projections demonstrated that future growth could be similarly offset by inequality. Insummary, growth, although beinga very important tool, is neither sufficient nor a necessary condition for poverty reduction. Policymakers must examine the reasons behind growth. In addition, growth could be offset, partly or completely, by inequality. Thus, a major policy concern would be for Turkey to address the persistence o f inequality while at the same time pursuingpolicies for highand sustainable growth. Turkey's continued commitment to structural reforms i s not only key for maintaining the hardwon macroeconomic stability but also to create an environment conducive for new domestic and foreign investments. Inflation-Poverty Relationship Highinflation acts a direct source of poverty by increasing income inequality insociety, Poor people generally have fixed incomes and under an inflationary environment, their real incomes are eroded through the mechanism o f inflation tax. However, for the non-poor, there i s a hedging option. High inflation rates increase uncertainty in an economy and returns from investment tools tend to be higher than inflation reflecting the implied risk premium. People using these hedging options can compensate the effects of the inflation tax and earn over and above the inflation rate through these high returns. On the contrary, poor people do not have savings to invest inthese hedgingoptions and they bear the burdeno f the inflationtax. InTurkey, between1994and2002, there was highinflationand, on average, positive growth. In this period, the Gini coefficient increased by 1 percent, which indicates an increase in income inequality. Poor segments o f the population were faced with the inflationtax, whereas non-poor people benefited from the effects o f inflation by investing heavily in government bonds with extremely high interest rates. This worsening in income equality counteracted any positive impact due from positive growth duringthis period. However, in2003 and early 2004, inflation and interest rates declined sharply due to the Central Bank's successful disinflation program and the government's strong fiscal stabilization policies supported by structural reform. Reduced inflation rates are likely to have had a decreasing effect on poverty rates inthis period. Inan environment of low inflation rates, the effects of the inflation tax can beminimized. Due to low inflation and reduced uncertainty, extra gains from investment tools, like government bonds with high profit margins, can be decreased. Thus, distortions in income distribution and the negative impact on poverty can be mitigated. To promote poverty reduction, the disinflation program should continue to be implemented, while fiscal prudence is maintained. Continued commitment to the structural reform agenda i s a key factor that will strengthen macroeconomic stability and enhance gains from the disinflation process. 19 PART 111: POLICY RECOMMENDATIONS A. MACROECONOMICPOLICYRECOMMENDATIONS AND GOVERNMENT REFORMPROGRAM The JPAR and the previous poverty assessment emphasize that the creation o f well-paying formal sector jobs i s the main lever for poverty reduction in Turkey, and that macroeconomic instability has severely limited Turkey's attempts to reduce poverty since 1987. The Government's reform program begins with implications for reducing income inequality through improved taxation, discussed below. First, though, the Government's action plan for creating macroeconomic stability and growth, as supported by the Programmatic Financial and Public Sector Adjustment Loan I11(PFPSAL 111),i s detailed. The main objective o f PFPSAL I11 i s to support implementation during 2004 o f the Government's financial and public sector reform priorities while ensuring that social programs are adequately hnded and increasingly better targeted. Key reform priorities in the financial sector include: (a) strengtheningthe regulatory framework for banking; (b) buildinginstitutional capacity at the Banking Regulation and Supervision Agency (BRSA) and the Savings Deposit Insurance Fund (SDIF); (c) restructuring and privatizing state banks; and (d) improving the corporate insolvency regime. Key reform priorities inthe public sector include: (a) deepeningo f structural fiscal policies in support of sustainable fiscal adjustment; (b) implementing public expenditure management (PEM) reforms covering budget planning and execution, financial accountability, and public liability management; and (c) strengtheningpublic sector governance including implementation o f the national anticorruption strategy and preparation of the civil service reform strategy. Priorities for social spending include: (a) adequate expenditure for health, education, and social protection in the 2004 budget; and (b) better targeting o f social protection. Medium-termprojections suggest that sustained implementationo f economic reform is necessary for Turkey to attain macroeconomic stability and growth. Under these projections, the economy is expected to grow by about 5 percent during 2005-2006. Specific factors underlyingstability and sustained growth include: (a) greater confidence in the policy framework; (b) improved macroeconomic stability and declining real interest rates-which would stimulate private investment and consumption demand; (c) stronger export performance-which would permit faster import and output growth; and (d) higher extemal inflows, including greater foreign direct investment. Under this scenario, fiscal adjustment would yield a permanent reduction in the public sector borrowing requirement from about 10percent o f GNP in2003 to 5 percent of GNP in2006. This would underpin the projected stabilization of the net public debt stock-to-GNP ratio below 60 percent o f GNP by 2006. The programmed fiscal adjustment would also support the projected decline ininflationto single digits by 2005. On the creation of macroeconomic stability and growth, the Government's program emphasizes the critical role o f structural reforms. Policy continuity i s essential to maintain the stability achieved after the 2001 crisis, and to preserve the growth and macroeconomic perfonnance o f the last three years. The first reform priority i s to accelerate bankingreform to consolidate the renewed trust in the sector and to preclude the possibility o f a systemic crisis. The program 20 emphasis has now moved from crisis management to broad-based financial sector development. The second priority area is the comprehensive reform o f the public sector to address the underlying structural causes of Turkey's macroeconomic instability, ensure medium-term fiscal sustainability, and increase the transparency and efficiency o f public expenditure management. The third priority area is continuation of vital reforms in the agriculture, energy, and telecommunications sectors to improve the climate for private investment in an effort to raise productivity, growth, and incomes. Many of the latter reforms also have a structural fiscal dimension. Turkey needs comprehensive financial and public sector reforms to definitively break with its history of inadequate transparency, public deficits, inflation, and financial instability. A quick retrospective highlights the structural weaknesses in both the public and financial sectors that contributed to macroeconomic instability and created the preconditions for the 2001crisis. Chronic public sector deficits led to the build-up of debt and the rapid expansion of the banking sector, which greatly outpaced the capacity of the regulatory framework. Macroeconomic distortions arising from fiscal imbalance created incentives for the banks to take on excessive risk, which was not curtailed by adequate regulation and enforcement. The absence of ample, accurate, and timely fiscal information-particularly the reliance on off-budget funds and quasi- fiscal obligations, including the so-called "duty losses" of the state banks-reinforced political and institutional obstacles to structural change and sustainable fiscal adjustment. Unhealthy interactions between inadequate fiscal and financial transparency compounded the risks to the economy, and helped ensure the build-up of a major systemic crisis following earlier cycles of crisis and aborted stabilization efforts throughout the 1990s. After undertaking urgent financial restructuring to address the impact on the banks o f the November 2000 and February 2001 crises, the authorities are making progress in correcting the underlying weaknesses in the financial sector. In parallel, institutional and policy reforms are being deepened to address systemic problems in the public sector, together with further structural measures to gradually underpin the fiscal adjustment. These reforms complement ongoing actions to improve the climate for private investment and promote growth. Rebuilding investor confidence and sustaining economic growth depend critically on the Government's ability to maintain the momentum of reforms to address the structural roots o f Turkey's chronic financial instability. Strong fiscal performance has been the cornerstone o f the economic program in the last few years. Fiscal gains were significant in2003, and the primary surplus rose from 4 percent of GNP in2002to over 6 percent ofGNPin2003, close to the programmed 6.5 percent target. Although the 2004 budget passed in December 2003 was consistent with the 6.5 percent primary surplus target, a sizable gap quickly emerged. The Government announced an above-inflation increase inminimumwages, and it also cut contribution rates for social security to reduce the additional costs to employers. In addition, the Government increased pensions by 21 percent, well above the inflation target. These initiatives, together with revenue shortfalls relative to the budget, created a financing gap o f close to 1.7 percent of GNP. The Government introduced a fiscal package in March 2004 to close the fiscal gap. The Government is implementing additional measures to meet the primary surplus target of 6.5 percent o f GNP for the consolidated public 21 sector-af which 5 percent of GNP is for the Central Government-set inthe revisedbudget for 2004. The 2004 fiscal package has two maincomponents: original budget measures and supplementary budget measures. To meet the 2004 primary surplus target, the authorities undertook a set of measures in December 2003 within the context o f the original 2004 budget. The special communication and transaction taxes became permanent, andthe surcharge o n income taxes, was extended for one more year. Motor vehicle taxes were adjusted upward to compensate the reimbursement of 2003 payments for an additional car tax (ruled unconstitutional). In addition, savings also came from cuts in investment programs, lower value-added tax (VAT) rebates for pensioners, and by eliminating vacancies when public sector workers leave their positions. The special appropriation6 has been eliminated. The second set of measures was introduced with the supplementary budget in March. The supplementary budget law cuts discretionary spending by 13 percent across all ministries. The Government also introduced measures to increase tax revenues by adjustingexcises of petroleumproducts, alcohol, and tobacco. The Government has adopted a strategy for tax policy reform that would bring Turkey's tax regime inline with best practice inthe European Union(Box 1). Box 1. Tax Policy Simplify and consolidate the indirect tax structure with a VAT with a standard rate and two low rates (1 percent for agriculture and 8 percent for other basic consumption goods). Introduce a unifiedspecial consumption tax (SCT) that will consolidate the current range o f excise and specific taxes into a single tax charged on a limited range of luxury goods. A threshold will be established for obligatory filing o f VAT returns. Earmarking of revenues will be abolished and all revenues from earmarked taxes will be brought into the budget. Simplify and consolidate the direct tax regime in line with OECD standards and international best practice. Convert tax rebate for wage earners into a tax credit to simplify administration and reduce tax burden on low- income wage earners. Minimize tax exemptions and exceptions. Harmonize real effective tax rates across all types o f financial instruments and maturities. Phase in the full indexation o f the corporate and business income taxes (real interest deduction to replace nominal interest deduction). Reduce the scope o f investment incentives, consolidate into a single investment allowance rate across sectors and geographic regions, and gradually eliminate withholding tax on these allowances. Make the incentive system more transparent and automatic. Overhaul the tax regime for Free Trade Zones. Terminate the corporate tax and payroll tax exemptions, Provide reduced corporate tax rate only on condition that separate subsidiaries are established and inward trading is below 15 percent o f sales. Adjust taxation o ffinancial leasing inline with internationally generally accepted accountingpractices. Implementation of the tax strategy i s progressing reasonably well. On the tax policy side, a unified special consumption tax (SCT) to consolidate a range o f excise and specific taxes into a single tax charged on a limited range o f luxury goods was enacted in June 2002. Implementing circulars for the SCT law were published in July 2002 and the tax went into effect in August 2002. Earmarking o f SCT revenues was eliminated, with retroactive effect for the 2003 budget, through a government decree issued inJanuary 2003. LnApril 2003, Parliament enacted the first legislative package under the direct tax reform designed to simplify andconsolidate the direct tax regime in line with OECD standards and international best practice. This legislation: (a) Special appropriation is a mechanism of converting earmarked revenues into appropriation duringthe year. 22 harmonizes tax rates on income from financial investments at the declaration stage; (b) simplifies and harmonizes the system of investment incentives; (c) reforms the system of income tax credits; and (d) simplifies taxation o f corporate earnings and dividends. K e y provisions o f the legislationbecame effective immediately, and the remaining articles became effective inJanuary 2004. Subsequently, a second package of direct tax reform legislation was enacted in January 2004 to minimize geographic, sectoral, and other investment incentives-including rationalizing the benefits in Free Trade Zones. The Government is strongly committed to addressing the problem of taxation of unrecorded income, and will prepare secondary legislation to require that all large financial transactions take place through the bankingsystem. Implementation of the Government's tax reform strategy is crucial for many reasons, not the least of which would be to address Turkey's highlevel of income inequality. Decomposition and simulations demonstrated that by addressing inequality, the poverty reduction effects of growth could be substantial. A transparent and progressive tax regime would enable Turkey to achieve other economic reform goals, such as expenditure targets for the social sectors, which in turn would help address the access issues documented in the health and education chapters of the joint volume. Turkey's continued commitment to structural reforms is not only key for maintaining the hard- won macroeconomic stability but also to create an environment conducive for new domestic and foreign investments. To promote poverty reduction, the disinflation program should continue to be implemented, while fiscal prudence is maintained. Continued commitment to the structural reform agenda is a key factor that will strengthen macroeconomic stability and enhance gains from the disinflation process. B. HEALTH POLICYRECOMMENDATIONS Turkey's current health care system is characterized by marked inequalities in access to and use of health services. These are driven by skewed distribution o f health care service provisions and medical personnel in favor o f the urban areas, particularly in the western parts o f the country; public spending on health care, which is skewed toward the upper-income groups; and a health insurance system that leaves an estimated one-third of the population without coverage. Those without insurance coverage are predominantly in the lower-income groups. The need to cover health care expenditures with out-of-pocket payments by all those who do not benefit from insurance coverage in turn leads a significant share of the population to forego health care, even when they are critically ill.Lower-income groups are significantly more likely to forego health care than upper-income groups. Coupled with marked inefficiencies in service provision, the system of unequal access and use has resulted in health outcomes that are not commensurate with Turkey's income level. Reforms in four areas could substantially help improve lower-income group access to health care, while contributing to an improvement of the country's health outcomes. These include: (a) expansion o f health insurance to assure better coverage o f low-income households; (b) undertaking steps to assure that disadvantaged areas with poor health outcomes have access to professional health care services; (c) strengthening the quality of primary health care services, 23 particularly in underserved areas; and (d) strengthening quality and outreach o f maternal and child health services and essential preventive care. Improve Health Insurance Coverage of Low-income Groups. Almost 60 percent of those living below the poverty line are not covered by health insurance, while HCIS data show that insurance coverage is an important determinant o f health-care-seeking behaviors. Therefore, expansion of health insurance to accord better financial protection to low-income groups would be a key step toward improving use o f health care by the poor. Introduction of universal health insurance is a core component of the Government's recently launched health sector reform program. The reform program will consolidate the four existing insurance schemes and expand coverage to those who have no financial coverage. The new universal insurance scheme is to be based on the principle of solidarity and risk pooling, with the State making premium contributions on behalf of those unable to do so. A key implementation challenge will be to assure that government contributions will indeed be targeted to those most in need, while minimizing errors of exclusion. Inthis context, options o f linking qualification for government- paid health insurance contributions to targeting mechanisms used for other social assistance benefits should be explored. To the extent that the planned universal health insurance fund will have various semiautonomous regional branches, funding equalization to ensure branches in lower-income regions with poor health indicators are adequately resourced should be based on regional demographic, health, and income profiles. Reduce Regional Disparities in Service Delivery. Supply-side issues are an important contributor to l o w health care use in rural areas in general, and in Southeastem and Eastern Anatolia, in particular. Posting and keeping physicians and other medical staff in isolated areas where many ofthe poor live has been a long-standing problem. While recently introduced salary supplements for postings in those areas may help somewhat alleviate the problem, it is unlikely that these measures alone will suffice to assure an adequate supply o f medical personnel inhigh- poverty areas. Inaddition to further incentive mechanisms (such as postings inplaces of choice after postings in shortage areas, and admission to high-demand specialization programs), consideration should also be given to establishment o f mobile health teams that visit isolated areas on a regular basis from their base indistrict capitals. Improve Quality of Primary Care. The perceived low quality of primary care leads the vast majority o f patients to self-refer to specialized, mostly hospital-based, care. This can lead to substantially higher direct and indirect costs for care seekers, and thus increased access barriers, particularly inrural areas. Circumvention o f primary care is also more likely to result inthe lack of continuity of care, which in turn leads to reduced effectiveness and use of care. Therefore, efforts to improve the quality of and the population's confidence in primary care, by ensuring that a greater range o f health needs can be effectively accommodated at the primary care level, could help reduce access barriers to health care. In this respect, planned efforts to introduce a family-medicine-based primary care system can be expected to have a positive impact on lower- income group access to and use of care. Inremote areas, efforts to deepen the technical capacity of health care personnel to allow for a broader range o f and more effective service provision at the primary care level would need to be complemented by measures that assure local availability of basic diagnostic services and essential drugs. 24 StrengthenMaternal and Child Health and Preventive Care. The continued high levels o f maternal and infant mortality in many areas o f Turkey calls for concerted efforts to improve the quality and use o f maternal and child health services. The fact that at least one-fifth of all births inthe East and Southeast continue to be unattended by any trained medical personnel (Table 2) and that up to one-third o f women do not receive any prenatal care,' point to serious shortcomings inthe Turkish medical system inthis area, with both supply and demand factors at play. Similarly, the continued marked regional variation in vaccination coverage points to serious shortcomings of basic preventive services, with large externalities. While recent data to ascertain the correlation between poor health outcomes and household income are not yet available,* earlier DHS data point to an expectedly strong correlation between health outcomes and household welfare (World Bank, 2003). Therefore, overall efforts to strengthen the quality and accessibility of primary care should integrate targeted efforts to improve access to and use of maternal and child health services and basic preventive care to reduce marked regional and intrahousehold variations inhealthoutcomes. The Government's recent Urgent Action Program and its sectoral spin-off, the Program for Transformation in Health, released in2003 and supported by the recently approved World Bank Health Transformation Project, target far-reaching sectoral reforms aimed at improving sectoral efficiency and increasing equity in access to health services. This is to be achieved through the separation of financing from service provision, introduction of financial and managerial autonomy ofpublic health care providers, more effective delivery of primary health care through the introduction of family medicine and, with particular importance to lower-income groups, and introduction of universal health insurance. The Program is expected to be fwlly implementedby about 2011, and will require major structural changes in the sector. Aside from the drafting of legislation to support these fundamental structural reforms, initial steps to eliminate access to health facilities based o n insurance affiliation have already been taken.' Similarly, important initial steps have been taken to encourage medical personnel to accept postings in underserved areas. While challenging on many accounts, the much-needed reform program, if implemented as envisioned, will indeed provide a technically sound avenue to substantially improve equity in access to health care. This in tum can be expected to effectively improve use o f health services by lower-income groups, which currently forego care. Together with planned efforts to improve the quality of care (introduction of family medicine, emphasis on improved matemal and child health services) they have the potential to help bringTurkey's health outcomes more in line with those of other OECD and European Union accession countries. As reform implementation progresses, it will be important to monitor its impact on access to care, particularly among the low-income groups and inthe more disadvantaged regions. 'Figures on lack o f prenatal care date back to the 1998 DHS, and may have improved somewhat since then. Data from the 2003 DHS, expected to become available during the second half o f 2004, will shed more light on accessibility and use of prenatal care, delivery, and child health care, including regional variations and variations across income groups. DHS 2003 data are expected to be available during the second halfo f 2004. Since early 2004, the distinction between SSK and MOH hospitals and health centers has been removed and patients are free to visit their facility o f choice. 25 C. EDUCA TION POLICYRECOMMENDATIONS Spendingon Education The 1997-1 998 education reform has been extremely effective in increasing the schooling attainment o f children coming from poor households. In 1994, wealthier households benefited more from public subsidies to basic education. Following the 1997 education reform, the distribution o f public spending on education at the basic education level became roughly flat across household income quintiles. The distribution o f public funds across schooling levels now favors basic education, which is another positive trend since the poor are more likely to benefit from these funds, and social returns to schooling are higher for early levels. At the secondary school level, too, there has been some improvement inthe distribution o f public funds, but there i s much room for further pro-poor shifts (which would occur ifpoor children's secondary school enrollment increases, as was the case with primary school enrollment). Only after increases in education spending associated with the financing o f the 1997 education reform did Turkey's public spendingon education as a percentage o f GDP become comparable to the spending patterns observed in countries o f similar levels of economic development. In order to ensure long-term economic growth and equity, there is a need to maintain public education spending levels of at least 4.25 percent o f GDP (which i s in fact a target stipulated in various Ministryo fNational Educationpublications). Enrollmentin BasicEducation The gross enrollment ratio for compulsory 8-year basic schooling i s 97 percent. Parental schooling is identified as a very good predictor o f enrollment. Enrollment rates for the basic education cycle could increase even hrther if children o f parents with no schooling (the demographic group most likely to be out o f school) were encouraged to enroll in school. Turkey's Conditional Cash Transfer program already provides financial subsidies to extremely poor families enabling these parents to keep their children enrolled in school. While this poverty assessment does not include a formal assessment o f the effectiveness o f this program, in principle, such targeted incentive schemes have been shown to be essential in ensuring the participationo f children from the poorest families to attend school. Enrollmentin SecondaryEducation This report demonstrated the importance o f the availability o f secondary schools to increase enrollment. But it also demonstrated that general secondary schools operate with significantly higherstudents-per-teacherratios than vocation schools. Thus, a first priority shouldbe ensuring optimal usage o f the physical infrastructure o f existing secondary schools. This alone may still not be sufficient, however, and thus there may be a need to proceedwith buildingnew secondary schools. The children who were found to be less likely to enroll in secondary school have similar characteristics to those in other countries: the poor, females, those with uneducated or absent 26 parents, those who reside in rural areas, and those who do not have a secondary school in their residential area. The magnitude o f the impact o f each o f these characteristics on enrollment is striking, as demonstratedby the analysis o f household survey data. The important point is that if compulsory schooling is further increased to 12 years, as mentioned in some Government planning documents, the beneficiaries would be almost exclusively the vulnerable groups listed above. Nonetheless, the prerequisites to such a move include raising efficiency inthe utilization of existing secondary schools and perhaps establishing new secondary schools. ParentalViews on the Quality of Schooling The parents o f poor children are more likely to report problems with schools, and less likely to be satisfied with the quality o f education their children receive. In 2001, the most commonly expressed problem with education was the lack o f books and supplies (29 percent o f the poorest quintile of households expressed this problem). Since then, the Govemment has acted on this issue by providing textbooks free o f charge to all primary school students. As a result, it could be that a major problem has already been resolved. N o w attention needs to shift to other main complaints from poor parents that emerge from the survey: namely the condition o f school facilities, poor teaching, and even a lack o f teachers (especially inrural areas). Tertiary Education The most effective way to ensure better representation o f poor children in tertiary education is through investments intheir schooling at earlier levels o f education. Interventions that come into play later in the education cycle are likely to be less effective. In fact, if high quality (free) primary and secondary schooling were made available to poor children, then perhaps the main intervention needed at the postsecondary education level would be to ensure poor students' access to credit and scholarships. Nonetheless, raising the quality o f primary and secondary education will not lead to significant increases in the enrollment o f the poor in tertiary education, in the short to medium term. Therefore, other interventions should be considered. Assuming that the centrally-run university entrance examination system (which has a number of positive features) is not changed, the Government may wish to design interventions to raise the performance of qualified poor children in these examinations. As discussed in the main chapter, wealthier children are much more likely to receive private courses (often through "dersane " attendance) specifically designed to improve scores on university entrance examination. Inthis context, policy makers may consider providing govemment subsidies to poor children to enroll in dersane-type university entrance examinationpreparationcourses. D. LABOR MARKET POLICYRECOMMENDATIONS The analysis o f the labor market demonstrated two key findings: the extraordinarily low rate of female labor force participation in Turkey and the strong link between informal and agricultural employment and poverty. It is difficult for policy makers to do much about the low rate of female labor force participation in the short run, but certainly implementing the reforms o f the 27 education system and the conditional cash transfers o f the Social Risk Mitigation Project (SRMP) should help in the future for more educated women to participate in the labor market. Also, the literacy training and micro-project endeavors under the Local Initiatives component of the SRMP should assist inthis long-term goal. On the linkage among informal and agricultural employment and poverty, here the best policy recommendationis for Turkey to stay the course on macroeconomic reforms, which have already led to significant real economic growth. As the economy grows and develops, more formal and better-paying jobs will emerge from that growth, andthese will provide the avenue for Turkey's 27 percent poor to find better jobs that will enable them to escape poverty. E. SOCIALPROTECTIONPOLICYRECOMMENDATIONS Recommendationsfor Reformof Pensions The immediate issue in the Turkish pension system i s the high and completely unsustainable level o f deficits requiring financing from the Treasury. The current high deficits and their rapid rise after an initial period o f decline following the 1999 reforms highlight the need for deeper parametric changes than hadbeen enacted in 1999. Potentialparametric changes include: 0 A rise in the retirement age, bothinthe short andlongrun. Currently, women mayretire as young as 41 and men as young as 43, but even inthe long run, the retirement ages are expected to stabilize at 60 for men and 58 for women. The current retirement ages are the youngest anywhere in the world and even the long run ages will look young relative to life expectancy when they have been finally reached. The system should aim for a life expectancy o f 15 years after retirement, with legislated increases in retirement age to match increases inlife expectancy andmove to this point as quickly as is legally feasible. 0 A rise in the number of years of contribution required before receiving a benefit. Currently, SSK requires only 19.4 years o f contribution to receive a full pension with a reduced pension available with only 12.5 years, while the other systems all require 25 years for men and 20 years for women. Internationally, full pensions are usually available after 25 and 40 years o f service, although usually the generosity o f the pension available at 25 years i s significantly less than what is available in Turkey. Reduced pensions for those unfortunate enough to have fewer numbers o f working years need to be available, but these need to be provided on a prorated basis, unlike the current system which gives the bulk o f the pension for the first years o f contribution, reducing the incentive for further contribution. 0 A reduction in the benefit accrual rate is also required to bring Turkey in line with other countries. Currently SSK pays 3.5 percent of wages in pension for the first 10 years o f contribution and then 2 percent per year for the next 15 years, while Emekli Sandigi pays 3 percent per year. The international averages are between 1 and 1.5 percent per year and even these are not fully affordable inall countries. 28 e A move to wage indexing the pensionable salary in lieu of the current valorization by nominal GDP growth would also be preferable. As Turkey moves, based on the 1999 Social Security Law, toward using career average salary as the basis for the pension, how the early wages are revalorized in creating this average becomes a critical factor in determining the value of the pension and the cost of pension payments. The 1999 law had specified GDP growth, which is historically higher than wage growth as the indexing factor. As a result, individuals who had earned the average salary throughout their career wouldhave a pensionable salary higher thantheir working wage. This bonus needs to be eliminated byusing wage indexation or some close approximation. e Maintainingprice indexation ofpensions as dictated by the 1999law is also important to containing costs. Prior to the 1999 law, Turkey had a tradition of ad hoc increases in pensions. The Government disregarded the 1999 law in 2004 and announced arbitrary increases inpensions once again, which were above the level o f inflation and raised costs notjust for this year, but also for all future years. e Eliminating and avoiding supplementary payments to pensioners. Prior to 1996, the Turkish Government had provided supplementary payments to pensioners, which became higher than the earned pensions breaking all links between contributions and benefits. These were frozen innominal terms in 1996, but resurfaced under a new name in 2003, social support payments. Politically, there seems to be a tradition that instead of reacting to low pensions by contributing for more years or by correctly declaring earnings, pensioners use the political process to exact higher pensions, thereby undermining the pension systems and incentives to contribute. a Maintaining or reducing the contribution rate. Despite the fiscal issues, raising the contribution rate in Turkey is not an option. As the fiscal situation improves, lowering labor taxes will serve to improve the efficiency of the labor market. Besides the parametric changes discussed above, the Turkish pension system is also plagued by being fragmented into four separate benefit schemes administered by three different agencies. Two of these four schemes retain the older, pre-1999, more generous benefit structure, resulting in large transfers from the state inparticular to the civil servant scheme. There is an issue of fairness as to why the state as a whole should favor one group over another as well as the administrative complications involved with workers who contribute to more than one scheme duringtheir working careers andof ensuringthat workers are indeed complying with the law and submitting contributions to one o f the four schemes. Thus, another recommendationis: a Unification of thepension schemes,both interms ofbenefit structure andinterms of administrationto improve equity and administrative efficiency. Finally, there i s still a large uncovered group of the elderly in Turkey. While the social pensions administered by ES do provide some safety net for the many elderly without an income, it would be beneficial to integrate this special program for the elderly with the other social assistance efforts in the country to try and reduce administrative costs and increase equity by using the same means testing methodology used by other social assistance programs. However, the social 29 pension administrator needs to also be aware o f minimum and average levels of contributory pensions to avoid providing such a generous noncontributory benefit that people choose to not contribute to mandatory schemes. e Integration of socialpensions with other social assistance benefits. The system for evaluating whether a person is eligible for a social pension i s separate from the systems, which determine eligibility for other types of cash benefits. There is no inherent reason that all social assistance benefits not be evaluated together and provided to families as a package, particularly given that the elderly poor mayreside within larger poor households. Recommendations for Reform of Social Assistance Despite recent improvements, Turkey still has a fragmented, under-funded and inadequate social assistance and social services system. Social assistance, interms o f cash and in-kindpayments to the poor is managed mainly by the Social Assistance and Solidarity Encouragement Fund (SYDTF) and its 931 affiliated Social Solidarity Foundations (SYDVs). However, a variety o f other programs are managed by institutions as diverse as ES (non-contributory pension for those over 65 as discussed above), the Ministries of Defense (for the families of conscripts), National Education, General Directorate of Youth and Sports, Prime Ministerial Office for the Disabled, municipalities and the State Social Services and Child Protection Organization (SHCEK, as well as a plethora o f minor programs under other agencies). Moreover, the SYDTF itself suffered from the lack o f a hlly concretized structure that was not rectifieduntilthe passage ofLaw 5263 on December 9,2004. Expenditures through the SYDTF, which until 2002 were fully funded through an extra- budgetary fund (EBF) receiving specified percentages of a diverse array o f government revenue flows, have fluctuated between 0.19 and 0.32 percent of GNP. This is very low by comparator standards-in many OECD countries, child benefits alone account for over 0.70 percent of GNP, with total social assistance rising sometimes to 2 percent o f GNP. Only in recent years have expenditures stabilized closer to 0.30 percent of GNP, with the encouragement of the World Bank and through the allocation of budgetary transfers (in 2004 reaching TL 200 million). By any standards, this is barely adequate. Under the medium to long term, with health expenditures o f the SYDTF transferred to universal health insurance (UHI), but the old-age social assistance pension transferred to the SYDTF, expenditures should be maintained at least at 0.60 percent} of GNP in order to have any meaningful impact on reducing poverty and establishing a robust social safety-net. Inthe short to medium term, expenditures should rise to and be maintained at 0.35 percent of GNPpending the necessary restructuring of expenditures. The Government, with the support o f the World Bank under the SRMP, has already begun the process of strengthening social assistance programs. This includes a substantial improvement of targeting through the use o f a proxy means test, the introduction of new poverty-focused programs to break the inter-generational cycle of poverty such as the Conditional Cash Transfers (CCT)-targeted to the poorest on condition that children are kept in school and that pregnant mothers and pre-school children attend well-care medical services, improved programs to lift the poor permanently out o f poverty (thereby not becoming long-term social assistance recipients) 30 through Local Initiatives providing support for income generation opportunities, improved robust and regular poverty monitoring and evaluation (M&E) through the State Institute o f Statistics, andgreater attention to M&Eo f social assistanceprograms overall. Despite these improvements, the overall system could still be much improved. Social services per se are weak andunder-developed. SHCEK, whilst providing good services at the provincial level in some provinces, has a spotty record, and fails to exercise an effective strategic policy- making and quality control role at the central level. Service provision is not well coordinated with other service providers (e.g., municipalities) and is frequently supply rather than demand- driven. The draft Public Administration Framework Law moreover appropriately transfers responsibility for service provision to the local level, designating the policy formulation, norms derivation, quality control and M&E functions to a reconstructed central level. Strengthening and reforming the social assistance and services system thus necessarily has the following elements, all o f which must be encompassed in framework legislation. Restructuring the system. This should provide for: (i) formulation, norms establishment, policy quality control and M&E for all population groups (children, youth, working poor, gender- related, elderly, disabled) at the central level in one single agency/ministry which would reduce the number o f central agencies; (ii)a single agency for determination o f eligibility and payment of cash benefits, namely the SYDTF-streamlining the system and enhancing efficiency, itself strengthened by a formalized legal structure; and (iii) decentralizing service provision to make it more responsive to local demands. Financing the system through a Government commitment to at least a minimum level of expenditure of 0.35 percent o f GNP. The quality o f Turkish fiscal adjustment leaves muchto be desired as key expenditures in the social area and public investment have been cut in order to finance recurrent expenditures. Targeting. The targeting o f social assistance can be significantly improved by using the same proxy means-test targeting used for the conditional cash transfers, but with a different cutoff point, depending on the desired number o fbeneficiaries. Improved targeting of social assistance is a reform being supported under the PFPSAL 111. This proxy means-test targeting system could also be used for fee waivers for health insurance. Expanding CCTs. Targeting o f social expenditures can definitely be improved as recommended above, and the fiscal savings from this would be well directed toward expanding CCTs, as well as any savings from future adjustments o f the public expenditure program or gains from sustained economic growth. The number o f beneficiaries o f CCTs was kept to 1.6 million children (and recently expanded to 7,000 pregnant women), or less than 2 percent o f the population. However, it is clear that there are many vulnerable, who, while not extremely poor, are still quite poor and who would benefit from keepingtheir human capital from eroding by an expanded CCT program. Additionally, Turkey could consider external financing for an expanded CCT program. For example, USAID has recently agreed to co-finance the existing CCT program, and a bilateral developmentbank finances many CCT programs inLatin America. 31 REFERENCES Milanovic, Branko, and Sholomo Yitzhaki. 2001. "Decomposing World Income Distribution: Does the World Have a Middle Class?" Policy Research Working Paper No. 2562. Washington, D.C.: World Bank. Ministry of National Education. 2001. "The Turkish Education and Developments in Education. " Ankara, Turkey. World Bank. 2000. "Turkey: Economic Reforms, Living Standards, and Social Welfare Study." ReportNo. 20029-TU. Washington, D.C., January27. . 2003. "Turkey: Poverty and CopingAfter Crises. ReportNo. 24185-TR. Washington, D.C., July 28. 32 ANNEX Table 1. Inequalityas Measuredby the Gini Coefficient;All CountriesIncludedin the Sample(Rankedby $PPPIncomeLevel) I Pooulation Gini I ZncomdExpenditures Estonia 0.383 I Lithuania 0.369 Guyana 0.490 Malaysia 0.463 Ireland 0.284 ~ Austria 0.472 Israel 0.347 Chile 0.564 Italy 0.306 Belgium 0.246 Taiwan 0.293 Australia 0.345 U.K. 0.354 Sweden 0.249 Netherlands 0.311 South Korea 0.310 Finland 0.226 Cyprus 0.297 Germany 0.294 France 0.326 0.246 Between-Country Gini 0.498 Within-Country Gini 0.161 World Gini 0.659 Source: Milanovic and Yitzhaki (2001) 33 Table 2. Per CapitaGDP and Poverty Below National 34 Below National 34 35 n.a. 36 UnitedKingdom 26,150 17 n.a. UnitedStates 35,750 12 n.a. -- n.a. = Not available. Sources: GDP and National Poverty are taken from Global Development Finance; and World Development Indicators; and World Bank Central database (2004). USSl PPPper personper day is taken from the World Bank website: http://www.worldbank.org/research/povmonitor/. 37