WP5; ?36 POLICY RESEARCH WORKING PAPER 28 35 A Decade of Fiscal Transition Asad Alam Mark Sundberg The World Bank Europe and Central Asia Region Poverty Reduction and Economic Management Sector Unit May 2002 I POLICY RESEARCH WORKING PAPER 2835 Abstract Transition literature has emphasized stabilization and * Reduced spending on government transfers enterprise restructuring. Both cross-country analyses and contributed to a sharp increase in income inequality in country-specific studies have tended to focus on fiscal the CIS. stabilization and its indicators, highlighting the * Fiscal risks increased during the transition. importance of quantitative fiscal adjustment to * Initial conditions allowed Central European and stabilization outcomes. Less attention has been paid to Baltic countries to maintain higher expenditures, the qualitative dimensions of fiscal adjustment in which may have contributed to their faster economic transition. recovery and political support for the reforms. Alam and Sundberg take stock of the extent to which The authors argue that the challenge today for fiscal adjustment has occurred during the first decade fiscal policy in these countries is to facilitate the of transition in both qualitative and quantitative transition-particularly in reallocating resources from dimensions. They define quality as the extent to which: large state-owned enterprises to new small and (1) pro-growth expenditure essential for creating future medium-size firms, and providing priority public economic and social assets are maintained; (2) pro-poor services and targeted transfers to assist those expenditure, such as poverty-targeted transfers, necessary adversely affected by transition and reverse the to ensure income for the poor and vulnerable are deterioration in social outcomes. The interplay adequately provided; and (3) fiscal risks, impinging on between fiscal policies and institutional arrangements both expenditure and revenue, are managed through is increasingly important as transition economies transition. embark on their second decade of reforms. In The authors conclude that while the quantitative particular, incentives embedded in the institutional magnitude of the fiscal adjustment was dramatic, the arrangements for fiscal management needs to be quality of this adjustment has compromised the social strengthened so that policies, resources, and outcomes and economic objectives of transition, particularly in the can be better aligned, and the fiscal adjustment is Commonwealth of Independent States (CIS). They draw consistent with qualitative considerations. four main conclusions: * Investments in public services fell in both absolute and relative terms. This paper-a product of the Poverty Reduction and Economic Management Sector Unit, Europe and Central Asia Region-is part of a larger effort in the region to understand economic transition in former centrally planned economies. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Alison Panton, room H4-156, telephone 202-458-5433, fax 202-522-2751, email address apanton@worldbank.org. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at aalam@worldbank.org or msundberg@worldbank.org. May 2002. (27 pages) The Policy Research Working Paper Series disseminates the findings of work in progress to encowrage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Research Advisory Staff A DECADE OF FISCAL TRANSITION by Asad Alam and Mark Sundberg* * The authors are senior economists at the World Bank. Part of this work was developed as background for the 2000 World Bank study on Making Transition Work for Everyone: Poverty and Inequality in Europe and Central Asia and the 2002 World Bank study, Transition: The First Ten Years-Analysis and Lessons for Eastern Europe and the Formner Soviet Union. The authors acknowledge with thanks commnents and inputs from Marcelo Selowsky, Pradeep Mitra, Christine Jones, Robert Conrad, Ricardo Martin, Michal Rutkowski, Kyle Peters, Lev Freinkman, Leila Zlaoui, Brian Pinto, Carlos Cavalcanti, Deborah Wetzel, Vinaya Swaroop, Sudarshan Gooptu, and various participants at World Bank and IMF seminars where this paper was presented. The authors are grateful to Victor Gabor and Afsaneh Farzin for help with data. I. INTRODUCTION The emphasis in transition literature has been on explaining the growth performance of the transition countries of Europe and Central Asia. Economic performance among these economies has differed widely and most of the analyses have centered on explaining these differences in terms of the indicators of stabilization, structural reform, and initial conditions.' At the start of transition, fiscal imbalances quickly emerged as a key challenge to macroeconomic stabilization, and were an inevitable result of lost revenues for transfer-dependent states of the former Soviet Union, the general collapse in incomes with concomitant loss in revenues, and additional demands for expenditures. Several factors further exacerbated these fiscal imbalances: the collapse of traditional tax systems based on large public enterprises; the dissolution of the CMEA trade system, the 'plan', and the monobank;2 the price shock from liberalization and hyperinflation which increased expenditure demands on the budget; and the shift of economic activities into the shadow economy with the resultant loss in revenues. Globally, all countries face the challenge of fiscal adjustment, be it in response to unanticipated shocks, economic mismanagement, or long term structural changes in the economy. Fiscal adjustment is effected through a mix of measures to reduce expenditures, strengthen revenue mobilization, and improve resource allocation and efficiency, often through institutional and structural reforms. What is different about fiscal adjustment in transitional economies are the marked differences in initial conditions relative to other developing countries. There are at least four major respects in which these differences are pronounced: * The magnitude of the required fiscal adjustment necessary for macroeconomic stabilization was far greater, often requiring a reduction in expenditures by as much as 20 percent or more of GDP in the initial years, especially in the CIS countries. Moreover, this fiscal retrenchnient had to take place in an environment of major and protracted economic contraction in the whole region; * The composition of expenditures required shifting from direct provision of most goods and services, characteristic of state ownership and central planning, to more selective provision or financing of only public goods and services and ensuring distributive justice. Moreover, the provision of public services and housing, which under central planning was mainly through state-owned enterprises, had to shift to either explicit government financing, or private sector financing. Budgets had to adjust rapidly to the changing economic realities. This required a massive reorientation of the role of the state, and contrasts sharply from most developing countries where the state, even if dominant, has not had as pervasive an influence in the economy; * The instruments for revenue mobilization needed to be transformed from direct appropriation by controlling the terms of trade and through enterprise turnover taxes (which were unrelated to enterprise profitability or value-added, opaque, and highly distortionary) to a market-oriented tax regime. In contrast, most developing countries have existing tax regimes based on modem instruments of indirect taxation; and * The extremely weak institutional legacy of budget and expenditure management systems-with its lack of key budget institutions that would ensure the appropriate matching of the costs of raising revenues with the benefits of expenditure programs.3 See, for instance, Fischer and Sahay (2000), de Melo et al (2001) and EBRD (1999). 2 See discussion in Tanzi and Tsibouris (1999). 3 See discussion in Hagen and Harden (1996). Key budgetary and financial institutions which were underdeveloped or non-existent included a meaningful budgetary system or budget law, adequate financial intermediation or bankruptcy laws to direct credit away from failing state enterprises to areas of growth, a modem treasury system for managing the cash resources of the government and ensuring timely payments and financial reporting, and a modem tax administration and legal structure for tax assessment and collection. While these institutional weaknesses are similar to many developing countries, the legacy of ubiquitous state control and a political economy focused on total state control contrasts sharply with the initial conditions in developing economies. While early fiscal policy analyses of transition economies focused on quantitative adjustment, there has been increasing emphasis in recent years on institutional issues and qualitative dimensions of fiscal reform. In the wake of the initial massive income shock and loss of revenues, attention was necessarily given to adjustment through fiscal austerity, especially expenditure retrenchment. Little attention was paid to distributive and efficiency considerations of fiscal reform where, in the absence of massive external aid, counter-cyclical fiscal policies were not a feasible choice. Since the mid-1990s, attention is increasingly being given to qualitative dimensions of the fiscal adjustment, i.e. the restructuring of public expenditures to support private sector growth, and cushion the social costs of transition, and the management of fiscal risk. Fiscal risks include the risk posed by contingent fiscal liabilities;4 tax and payment arrears, tax offsets, and other non-monetary instruments which have been persistent, distort budgets, and undermine competitive restructuring of enterprises;5 the quality of public expenditure management and the need for developing treasury systems, improved budgeting, promoting performance orientation; the economic risks posed by falling educational and health outcomes; and the low levels of transparency and accountability in fiscal management. This shifting emphasis is already supporting a richer policy dialogue promoting the pursuit of economy, efficiency, and effectiveness of public expenditures. A recent and comprehensive assessment of qualitative aspects of fiscal adjustment in the transition economies of Europe and Central Asia, however, has not been undertaken. An early effort was made by Cheasty and Davis (1996) and more recently Tanzi and Tsibouris (2000) and Gupta et al (2001). Cheasty and Davis recorded the then uneven progress on fiscal adjustment by the states of the former Soviet Union, often exogenously forced by inelastic financing. They recognized that the sharp reduction in deficits had contributed to stabilization but had also resulted in sharp misalignment of expenditure priorities and the expansion of unorthodox and disorderly budget procedures.6 This had compromised the quality of fiscal outcomes, caused major expenditure inefficiency, and showed little sign of significant budget restructuring. Tanzi and Tsibouris provide an overview of the fiscal reforms, including institutional reforms in tax policy, tax administration, and expenditure management, but they do not discuss the patterns, outcomes, or 'quality' of fiscal adjustment. Gupta et al assess the changes in the size and scope of government in transition economies. They conclude that while these governments have retrenched in terms of public expenditures in relation to GDP as well as in public sector employment as a share of the population, various indicators suggest that government size remains high even as the scope of government is too broad. While the paper does have some discussion of issues relevant for quality of fiscal adjustment, the focus of the paper is more on the quantitative aspects of adjustment and on the extent to which the scope of government has been aligned with market needs. 4 See Brixi, Ghanem, and Islam (1999), Brixi, Papp, and Schick (1999). 5 Pinto et al (1999) have analyzed the Russia situation is detail. 6 See Cheasty and Davis (1996). 2 Many of the conclusions of these assessments remain valid and relevant, though a more nuanced understanding of fiscal adjustment is needed. The challenge for fiscal policy has been to facilitate the economic transition, in particular the reallocation of resources from state-owned and typically large enterprises towards restructured and new small and medium firms, while continuing to provide priority public services and targeted transfers, and helping ensure that those adversely affected by the transition are taken care of.7 While there is no doubt that fiscal policy remains central to transition, the qualitative dimensions of fiscal adjustment-ensuring the provision of basic public services, including social protection to the poor and the vulnerable, and minimizing fiscal risk while maintaining a sustainable fiscal deficit-have become paramount. Such an ordering of fiscal policy choices can only come about from the proper alignment of incentives in the institutional framework for budget management. This interplay between the institutional and policy aspects is increasingly important as transition economies embark upon their second decade of reforms. Indeed, evidence suggests that even for those countries more advanced in the transition process, strengthening the institutional framework for budget management is equally important. This paper seeks to take stock of the extent of fiscal adjustment undertaken by the transition countries in the turbulent initial decade of their shift towards a market economy, and discusses both quantitative and qualitative dimensions of this fiscal transition. The paper then draws some lessons to guide future policymaking. H. THE EXTENT OF FISCAL ADJUSTMENT Any discussion of fiscal data on transition economies necessarily has to highlight two challenges: data deficiencies (especially in the early years of transition) and the significant differences in initial conditions: * Data Issues: The absence of good government accounting systems and standard data definitions puts the quality of fiscal data at serious risk. An attempt has been made here to provide some degree of comparability and coverage across the region. The paper draws mainly on data collected by the IMF's Fiscal Affairs Department, supplemented in some instances by World Bank sources. Government expenditures and revenues refer to consolidated expenditures of central and sub-national governments, including extra-budgetary funds (such as pensions, road funds, and social insurance funds). Expenditures and revenues presented in the paper are on a cash basis and, therefore, exclude arrears as well as quasi-fiscal deficits (e.g. liabilities arising from the banking system) and may consequently overstate fiscal balances. Notwithstanding these deficiencies, and risking inevitable loss of precision, the statistical evidence nonetheless sheds some light on the nature of the fiscal transition. * Initial Conditions. Large differences in initial conditions inevitably impinged upon the fiscal options as well as the choices countries made. All of the transition countries started the 1 990s with a very limited private sector, an absence of well defined commodity or factor markets, hence the absence of prices acting to signal scarcity, and with a large state sector which dominated the economy and provided many public services-housing, schooling, medical care, etc.8 However, the initial conditions in the central European and Baltic (CEB) countries were more favorable for a rapid transition for several reasons. On average, the private sector in the CEB countries was very 7 See World Bank, 2002, Transition: The First Ten Years, for greater elaboration of this theme. 8 See de Melo, Denizer, Gelb (1997), Lazear (1995), Berg et al (1998), Havrylyshyn and Van Rooden (1999), Tanzi and Davoodi (1997). 3 small (Poland had the highest share at 20 percent of GDP in 1991) but still larger and better poised to fill the gap left by collapsing state enterprises than was true of the Commonwealth of Independent States (CIS) countries. A greater share of CEB trade was already directed towards West Europe, and the collapse of the CMEA trade arrangements posed a less severe shock to their economies. Moreover, selective reform of state enterprises had been initiated since the 1980s in some of the CEB economies (Hungary, Poland) softening management rigidities and dependence on the state. These differences help to partly explain the subsequent patterns of fiscal adjustment. Several stylized facts emerge from an analysis of the fiscal data and are discussed below: (i) All the transition economies went through a dramaticfiscal adjustment in aggregate terms (see Figure 1). The turnaround in fiscal imbalances has been especially remarkable for the CIS countries, which have reduced their average cash deficits from a high of 19 percent of GDP in 1992 to a surplus of 0.7 percent by 2000. The extent of this adjustment during this period is more than fives times that of the CEB countries whose average deficit was reduced from about 7 percent to 3.2 percent over the same period. Starting with much larger fiscal deficits, the CIS countries were compelled to make larger expenditure cuts. Between 1992 and 2000, the average reduction in expenditures in CIS countries was 23.1 percent of GDP as compared to 5.6 percent in the CEB countries (see Table 1). The picture is more compelling when one looks at the changes in primary balances since interest expenditures have been growing during this period. While the CEB countries have generally maintained balance on their primary budget, the CIS countries have drastically reduced their large deficits of over 10 percent of GDP in 1992 to a surplus of over 4 percent of GDP by 2000. Figure 1: Fiscal Contraction during Transition, 1992-00 Fiscal Contraction during Transition Overall Deficit 10 (5 -0 0 co 5 - 1992 1993 1994 1995 1996 1997 1998 1999 2000 Fiscal Contraction During Transition Primary Balance 9.0~~~~~~~~~~~~~~I 21.0 5.04, 1992 1993 1994 199 1999 1997 1998 199 2000 Note: The group averages are population-weighted Source: IMF and World Bank data; Authors' calculations 4 Table 1: Total Budget Revenues and Expenditures-change in percentage points of GDP, 2000 over 1992 Revenues Expenditures CEB -2.1 -5.6 CIS -3.7 -23.1 Note: All changes are based on population-weighted averages. Source: Author's calculations (ii) The drastic fiscal adjustment in the CIS countries was necessitated by a major revenue shock at the start of transition. For many of the CIS countries, independence from the Soviet Union also meant the loss of large fiscal transfers from Moscow, which further compounded declines in government revenues from recession and flawed tax systems with weak administration. While precise and comparable data on the extent of the loss in transfers from the former Soviet Union are not available, some country examples are illustrative. For instance, in 1992, both Uzbekistan and the Kyrgyz Republic lost transfers from Moscow which were equivalent to about 18 percent of GDP in 1991. Other causes of the revenue decline were the concentration of output loss in the traditional tax bases, revenue-losing tax reform, under- utilization of the revenue potential of the energy sector, and the monetization of the implicit subsidies.9 (iii) In the face of the sudden loss of control over state resources in the CIS countries, Governments had no choice but to sharply cut expenditures (Figure 2). For the CIS countries, aggregate expenditures fell from about 56 percent of GDP in 1992 to 33 percent of GDP in 2000; for the CEB countries, expenditures fell from an average of 48 percent in 1992 to 42 percent in 2000. In some cases, the expenditure cuts were dramatic: in Georgia, general government expenditure declined from around 36 percent of GDP in 1992 to just 12 percent in 1995 before recovering to25 percent of GDP by 1998. (iv) Since the 1998 financial crisis, fiscal balances have improved sharply in the CIS countries. Much of this is on account of the dramatic improvement in Russian finances, in large part due to the boom in energy prices, the resultant income gains, and the growth in tax revenues therefrom. Other energy exporting countries in the CIS have likewise gained from this favorable terms of trade development. But this also reflects "leaming from the crisis" which inculcated in policymakers a greater appreciation of the need to put public debt dynamics on a sustainable path and greater political consensus on this issue. This was reflected in marked changes in fiscal behavior as countries sought to restrain expenditures, manage extemal debt more prudently, including reduced exposure to interest sensitive short term debt, improve revenue collection, and strengthen budget management. The recent improved budgetary performance in the CIS countries marks a departure with the pattem of change in some of the CEB countries (e.g. Poland) where maintaining aggregate fiscal balance in the wake of growing public expenditures has become more difficult to manage. 9 See Cheasty and Davis (1996) and the references therein for a fuller discussion of the causes and consequences of the revenue decline. Gray (1998) discusses underutilization of energy revenue potential in the former Soviet Union and Baltic countries. Figure 2: Expenciture Reductions during Transition, 199200 60.0 - 55.0 - A I 50.0 ~ CEB 45.0 40.0 35.0 1992 1993 1994 1995 1996 1997 1998 1999 Exp 2000 Note: Averages are population-weighted Source: IMF, World Bank data; Author's calculations (v) By 2000 most countries had achieved manageable budget deficits in the 2-4 percent of GDP range, though there is significant variation across countries. While most countries are clustered in a narrow band, there are extremes--Russia's overall budget in 2000 posted a surplus of about 3 percent of GDP while Albania's posted a deficit of 9 percent of GDP (see Figure 3). While the genesis of these variations are of course different, recent years have shown a convergence towards moderate budget deficits. This picture is reinforced when one observes the strong fiscal positions of many countries as reflected in the primary budget balances. 6 on 0 % of GDP *000 boooo -~~~~~~~~AloM Albania cm~~~~~~~~~~~~~~~~lania Kyrgyz~ ~~~~~~~~~~~~~yy Rpbi 0 AAbania Armenia C~~~~~~~~~~Kgzec RepublIc 1 Urbkmeistan Athuania I Azerbaijan Sl~~~~~~~~~~~~~ova eulicv Uthuania ~ ~ ~ ~ ~ ~ ~ ~ ~~~dwi Siavak Republic~~~~~~~~~~~~~~~~l Latvia 020 Slovak cr 3 Georgia 9TI Estonia !M ueriin Belarus Tajikistan 0)Ubksa Poland Lalva Slovenia Azeibaja Georgia Ukraure Romania Eslorlia Ukraine Slovenlia Kazakhstan Ta"ldtan Hungary Bulgaria Moldova Beiatus Bulgaria Kazakttstan Macedonia, Macedonia, FYR Russian Russian Federatin (vi) Revenues have now stabilized though inter-country variations are significant. For the CEB countries, government revenues (population weighted) have stabilized at about 39 percent of GDP, only one percent of GDP lower than early transition levels; for the CIS countries, revenues fell from about 38 percent of GDP in 1992 to around 31 percent of GDP in 1999, and rose to 34% in 2000 due largely to increased oil prices. Country variations are marked (see Figure 4, with simple averages). The advanced market reformers, which more readily adapted to markets and had stronger economics links to West Europe, have more readily sustained high tax collection rates. In Croatia, Czech Republic, Hungary, Latvia, Poland, Slovakia, and Slovenia, tax to GDP rations were in the 40-45 percent range in 2000-comparable to tax collection rates in some advanced high-income countries, and probably too high to be sustainable, as suggested by Figure 5. (Belarus is the only other country in transition to have comparable revenue ratios). Many of the CIS countries have very low tax collections. In previously war-torn Georgia, for example, government revenues dropped to 7 percent of GDP in 1995 before recovering to about 16 percent of GDP by 2000. Similarly, Turkmenistan, Kyrgyz Republic and Tajikistan had revenues of about 14 and 17 percent of GDP in 2000. Figure 4: Revenue Mobilization in CEB and CIS Countries, 2000 60.0 & - U. 50.0 EF 40.0 30.0 - - - - - 20.0 - 10.0- - - - - - - Source: IMF; author's calculations (averages are unweighted). (vii) Expenditures show variations among countries and particularly between the CIS and CEB countries. Expenditure levels are typically higher in CEB countries, on account of higher social expenditures (discussed below). Expenditures vary from a low of 14 percent of GDP in Tajikistan to over 50 percent of GDP in Croatia and Hungary (Figure 5), approaching the higher levels found in OECD countries. For some of the poorer countries in the CIS, foreign aid inflows have helped maintain expenditures at levels higher that would have been possible. 8 Figure 5: Expenditure in CEB and CIS Countries, 2000 60.0 5 0.0 20.0 N 30.0 - - - - - - - - - - Source: IMF: author's calculations (averages are unweighted). (viii) The larger relative quantitative adjustment in the CIS countries also reflects larger 'imbalances' in expenditure levels relative to income level. Relative to their average income level, the size of government in the CIS countries reveals a far greater magnitude of expenditure cuts during transition to a market system. Figure 6 shows the average share of Government expenditure to GDP, relative to the log of per capita income. The trend line represents the regression of size of government on log of per capita income (PPP adjusted to correct for major exchange rate distortions) across a sample of developed and developing countries for which comparable fiscal data was available in 2000. The simple relationship shows the increasing size of Government with per capita income level (commonly called Wagner's law), around which there is considerable variation. Figure 6: Government Size and Income Level, 2000 -. 55 C, .~45 -GB0 e 25_ ~ 2. 5 - . . 5L E E @>1 15 0 5- 2.5 3 3.5 4 4.5 5 LOG of Per capita Income PPP based Trendline: Y=1.7 X - *.0, with R2=O.3 . Based on a sample of 49 developed and developing countries vwith comparable fiscal data. Source: World Bank, IM F, and Authors calculations Source: IMF; World Bank; Authors' calculations 9 The changes in the levels of expenditure in the CIS and CEB country groups for 1992 and 2000 (using simple averages) are also shown relative to the trend line. It is noteworthy that both the CIS and CEB countries were at expenditure levels comparable to OECD economies in 1992 (exceeding 45 percent of GDP), and show a sharp deviation from the cross country trend line. However, by 2000, their relative positions had diverged quite strikingly. The massive expenditure cuts in the CIS countries brought them to a point right along the regression line for their income level, at around 28 percent of GDP. The much more modest contraction in the CEB countries, however, shows them still well above the regression line, indicating the persistent large size of govemment relative to their income levels. This suggests that further reduction in the size of government in CEB countries may be warranted (particularly with respect to expenditures on the social insurance programs that may prove difficult to sustain), although this is fundamentally a question of socio-political preferences and sustainability. III. THE QUALITY OF FISCAL ADJUSTMENT Given the dramatic size of the fiscal adjustment, what can we say about its quality? Recent literature on fiscal adjustment has emphasized the quality aspects along three dimensions-the extent to which (i) pro-growth expenditures that are essential for the creation of future economic and social assets are maintained or protected, (ii) pro-poor expenditures, such as poverty-targeted transfers, necessary to ensure income streams for the poor and the vulnerable are adequately provided for; and (iii)flscal risks, impinging upon both expenditures and revenues, are reduced or, at least, contained. These categories are of course not mutually exclusive, and it is easy to think of ways in which pro-growth expenditures, for instance in rural infrastructure, can also be pro-poor by enabling better access to markets and reducing transport costs for the rural poor. Better targeting of pro-poor expenditures can help reduce the level of expenditures and strengthen fiscal discipline, thereby improving prospects for macroeconomic stabilization and growth. A better management of the fiscal risks ensures that risks of periodic bouts of macro instability with high inflation and economic and social costs-in which the poor typically suffer disproportionately, and which undermine growth prospects-are reduced. A. Protecting Pro-Growth Expenditures Three groups of expenditures are considered here: human capital investments in education and in health as reflected by levels of public expenditures, and capital expenditures on the creation and maintenance of the country's fixed capital stock. Such a nomenclature is necessarily limiting for it implies that public expenditures in human and physical investments are critical for growth. What is missed here are total expenditures in the economy in these areas which would include both public and private expenditures. Indeed, private expenditures on health, education, and physical investments have become more important in transition economies, and may well be offsetting any decline in public expenditures. Moreover, other forms of capital-such as social capital-may also be necessary components of a growth strategy. However, such a narrowing of scope is essential here in order to keep the discussion manageable and focused. (a) Public Expenditures in Education The fiscal adjustment in the region has taken place in part at the cost of cutting education services even as most countries have tried to protect spending on education.'° While comparable time series data is not available, given the large declines in output in all the countries, and the to In relative terms, education spending increased. See World Bank, 2000, ibid. Chapter. 7. 10 retrenchment in aggregate expenditures, the real value of expenditures on education fell. However, expenditure levels were maintained as a share of smaller GDP's in most countries." The cuts in real expenditure are reflected in deterioration in indicators of learning opportunity, access, quality, and coverage.'2 Though education outcomes are still much higher that those found in countries of similar income levels, some of the CIS countries show significant declines in enrollment rates for pre-school, basic education, upper secondary education as well as vocational and technical education though there has been some recovery in rates in some countries with the recent pick-up in economic growth. In the poorest countries-Moldova, Armenia, Georgia, and Tajikistan-enrollment rates fell by about 10 percentage points or more during the past decade. Essential expenditures on operation and maintenance inputs-such as on textbooks, school supplies, and building maintenance-were squeezed. For instance, in Ukraine, allocations for textbooks fell by 70 percent between 1995 and 1998, and capital maintenance dried up even as the number of teachers expanded. Poorer regions have been forced to bear a disproportionate share of the adjustment. For instance, in Georgia, reportedly 43 percent of primary and secondary schools in urban areas got textbooks for all children compared with only 27 percent in poorer rural areas. In Russia, where regions have responsibility over most educational funding, richer regions have been able to spend more on education while poorer regions have had to struggle to maintain the basic requirements, contributing to the rise in income inequality.'3 Declining public expenditures in education have also shifted costs to households. While this may be a more efficient solution, concem arises from the growth of informal payments for publicly provided services, and the burden this creates for the poor, limiting their access to education services. Despite budget pressures and falling enrollment rates, teaching staff has grown in many countries. In Russia the number of teachers expanded by 25 percent between 1989 and 1996, and in Central Asia, increases range up to 25 percent.'4 Student-teacher ratios are typically very low (for instance, 8.7 in Armenia, 9.9 in Azerbaijan, and 11.9 in Russia) and it is widely held that this can be increased without compromising teaching quality or learning outcomes. Using OECD standards as a benchmark, up to a third of the teaching labor force can be reduced. However, policy makers have typically chosen to cut less politically sensitive outlays on operations and maintenance, and have deferred necessary capital repairs and energy efficiency improvements, rather than take the politically difficult step of laying off teachers and other state employees. The real cuts in education expenditures have occurred even as cost pressures have remained high; little has been done to re-allocate towards more productive uses. High energy costs-which in some of the colder CIS countries absorbs 30-50 percent of expenditure outlays- is a significant cost burden on the sector. There has been only limited effort towards reducing energy costs through energy conservation incentives and investments in energy efficiency measures. Additional pressures on the public purse have come from other sources. In countries such as Georgia, Armenia, Tajikistan, and Azerbaijan, budgetary resources are inadequate to meet the fiscal requirements of universal education coverage. In almost all countries, rising tertiary education enrollments-which reflects the demand for higher human capital in transition economies-risks absorbing larger shares of education budgets unless tertiary education is financed through increasing cost recovery, reduction of unit costs, and private sector delivery. " Inplicitly, this means that part of the 'peace dividend' in reduced defense expenditures has gone towards reducing the burden of adjustrnent on education (and other public) services. 12 For a fuller discussion of these issues, see World Bank (2000), chapter 7. 3 See World Bank (1999). "4 Klugman (1999) and World Bank (1999). 1 1 The past practice of public financing and provision for these services is no longer viable.'5 Even as falling preschool and basic education enrollment rates have led to low utilization rate of schools and high maintenance costs, adequate savings have not been generated by consolidating schools and rationalizing teaching staff which can help generate savings to more selectively finance operations and maintenance requirements, including unmet demand for textbooks and school supplies. Notwithstanding the real cuts in education expenditures, levels of such expenditures as a share of GDP remain high compared to other countries of the same income level. At the end of the 1990s, both CEB and CIS countries maintained expenditures at around 4.6 percent of GDP (see Table 2). This compares quite favorably to low and middle income countries with average education spending of 2.9 percent and 3.5 percent of GDP respectively whereas high income countries averaged 4.8 percent of GDP, roughly the same as the CEB and CIS averages. This, of course, masks significant variation in expenditures across the region as public spending on education ranges from under 2 percent of GDP for Armenia and Georgia to almost 8 percent of GDP for Uzbekistan.'6 Some CIS countries, particularly those with the sharpest aggregate expenditure adjustments, were forced to cut their education outlays sharply, while the slowest reformers among the CIS group (Belarus, Turkmenistan, Uzbekistan) have sustained higher spending levels. Table 2: Selected Public Spending (% of GDP) Education Health Social Protection CEB 4.6 5.1 13.3 CIS 4.6 3.6 7.4 Europe & Central Asia 4.1 4.0 -- East Asia & Pacific 1.7 1.7 -- South Asia 3.1 0.9 -- Sub-Saharan Africa 4.7 1.7 -- Low Income 2.9 1.2 -- Middle Income 3.5 2.5 -- High Income 4.8 6.1 -- Notes: Simple averages have been used for the regional averages. Data is for most recent year which varies by country. Source: World Bank (b) Public Expenditures in Health Public expenditures in health also fell in real terms during transition with the decline in national incomes, but-like with education expenditures-most countries also made an effort to protect these expenditures in relative terms. Budgetary health expenditures as a share of GDP have been remarkably stable across the decade, with average expenditure levels in 1998-99 which were close to, or even above, expenditure shares at the beginning of the 1 990s (Figure 7). As a share of total public spending, health outlays have increased significantly over the decade. These expenditure levels are well above the average spending of low and middle income countries, reflecting the legacy of universal public health care coverage (Table 2). There is a wide disparity in public health expenditures, however, within the region, from below I percent of GDP in '5 The young demographic profiles of these countries imply potentially larger future cohorts for higher education. This highlights the growing importance of cost containment. 16 For more discussion and data, see World Bank (2000), chapter 7. 12 Georgia to over 8 percent of GDP in Croatia, exceeding typical OECD expenditure levels. Public health expenditures are clearly inadequate in several CIS countries where spending has been dramatically cut.'7 Figure 7 Health expenditures, 1990-1999 6.0 5.5 - cL 5.0 40 4-5 40 03.5 - - -*- _ -- EL 4.0 -&rtn 2.5 - 2.0 - 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Source: World Development Indicators Though transition countries have better health indicators in most cases than other countries of similar income levels-reflecting the universal provision of comprehensive health care services and high levels of education- disturbing evidence of worsening health outcomes has emerged.'8 Male life expectancy fell by around four years between 1989 and the mid-1990s in the Baltic countries, and by five years in Russia, Ukraine, and Kazakhstan. Immunization rates have fallen in several countries-and at disproportionately higher rates for the poor. Thus, in Kazakhstan, while only 40 percent of the children in the poorest quintile have been immunized, the equivalent ratio is 80 percent for the highest 40 percent of the population. In Russia, Ukraine, and Moldova, previously rare communicable diseases, such as tuberculosis and HIV/AIDS, have become serious threats to public health. With rising poverty levels, other public health problems are emerging. For instance, in parts of Central Asia, undemutrition in children under five years of age is leading to higher rates of stunting and wasting. Similarly, micronutrient malnutrition is causing higher rates of anemia among women of reproductive age, particularly in the poorer Aral Sea region of Kazakhstan and Uzbekistan. Moreover, declining public health expenditures shifts health costs to families which may discourage access to health services by the poor. While wealthy famnilies, particularly in urban areas, can afford side payments or private health care alternatives, the poor are often forced to sell assets, borrow funds, or forgo services altogether. In the Kyrgyz Republic, for instance, three- quarters of all patients in 1996 made informal payments and one-third of all patients seeking inpatient services had to borrow money. Similar figures are reported for Georgia, Tajikistan, and Ukraine. In Russia, data from 1997 suggests that 41 percent of all Russian patients could not afford to purchase required drugs and 11 percent could not afford any medical treatment. Similarly, 37 percent of pregnant women in Tajikistan could not afford pre-natal care, and almost one-third of births occurred at home, a break from past practices of hospital births. 17 More details are provided in World Bank (2000), chapter 7. 18 See World Bank (2000) Chapter 8, for further details. 13 For most countries in the region, even with high levels of public expenditures, the intra- sectoral allocation of health expenditures remain low. Historically, the region has emphasized investment in hospital care and large medical facilities, with far higher hospital bed-to-population ratios than in high income countries. But now, financing needs to shift away from the maintenance of this large hospital infrastructure and specialist care towards providing basic health services particularly to the poor, for whom access to quality basic healthcare has seriously deteriorated. Financial incentives for health services workers are poor, encouraging demands for side payments for services, including payment for drugs and medical supplies. Over-staffing is evident from the high health sector employment as compared with OECD averages.'9 Health sector employment in countries such as Russia, Ukraine, and Kazakhstan is around 6.5 percent of total employment while the OECD average is only 4.3 percent. In Russia and Ukraine, there are 4.5 and 4.3 physicians respectively per 1000 persons compared with an OECD average of 2.6. Wage bills in health care have consequently increased as other health expenditures have fallen. In Ukraine, the share of wages in health sector outlays increased from 36 percent to 46 percent over 1995-98, while the share spent on medicines fell from 12 percent to 7 percent. At the same time the real wages of health workers have plummeted to one-quarter of their 1992 levels. There also remains wastage of resources on half-empty facilities, poorly insulated buildings, or dysfunctional machines. Sixty percent of the health budget in Moldova is spent on utilities due, in part, to the large excess capacity of hospitals and other medical facilities. The infrastructure downsizing program of Georgia is an example of how a sizeable reduction in beds and sale of unused buildings and facilities has led to a consolidation of the infrastructure. Technical inefficiencies in health expenditures remain high and are reflected in high unit costs of treatment. One key source derives from long hospitalization stays. The average length of state in a hospital for acute care is 16 days in CIS countries compared with 6 days in Western Europe. Hospitals in many of these countries have incentives to keep patients for longer duration since this partly determines hospital budgets. Establishing norms for hospitalization, and removing perverse budget incentives, is needed. Moreover, there remains undue emphasis on hospital and specialist care though this is now shifting to outpatient and preventative care. Greater emphasis on outpatient services and preventive health care, such as infant immunization, prevention of infectious diseases, and reducing "life style" induced health problems, can reduce costs and free up resources. The shift from specialists to family medicine practitioners, occurring in much of the region, will lower costs. (c) Physical Investment Expenditures For most countries, overall expenditure cuts fall most heavily on public investment and discretionary current expenditures, such as outlays on operations and maintenance. Forced fiscal austerity in the transition countries has similarly led to shrinking investment in physical assets as a share of GDP for most countries. This has occurred against a background of rising investment needs during transition, and sharply deteriorating infrastructure. Data are not available for the early years of transition though some country case studies reveal serious declines (at least initially) and disparities in levels of capital expenditures. In Russia, capital expenditures of the overall government halved from about 7 percent of GDP in 1989 to 3.5 percent of GDP in 1999.2o In Ukraine, capital expenditures fell from 4.6 percent of GDP in 1994 to less than 1 percent in 1997.2' In Kazakhstan and Azerbaijan, public investments '9 See World Bank (2000). 20 World Bank, 2001, Russia: Improving the Efficiency of Public Investment Expenditures, Report No. 22693-RU. 21 World Bank, 1997, Ukraine Public Investment Review, Report No. 16399-UA. 14 had fallen to 2 percent of GDP by 1998-99. In some countries such as Estonia, public investment fell in the first two years of transition to 1.4 percent of GDP before rebounding to around 5 percent of GDP by 1999.22 Even as the initial stock of physical infrastructure was aging and in decline at the start of transition, the subsequent decline in public investments has led to further deterioration in the stock of physical capital and has deferred investments in new assets. With the exceptions of Belarus and Uzbekistan (not shown), the CIS countries generally have levels of public investment lower than those of the countries of CEB even though many of the latter are more advanced reformers and attract higher levels of private investment. Current levels of public investmnent are widely dispersed, as expected, from lows of under 2 percent of GDP in Kazakhstan and Tajikistan to highs of 11.6 percent in Belarus (see Figure 8) where the vestiges of socialist state control remain most entrenched. However, these figures may not fully capture off- budget investment outlays which may be significant in some countries. IFAgutre a: Capital Expandituroo In Transition Countries, 1999 (or most rocont year) _1'3, a ii s la n .Sl ovent -Sovak Reublic .># . . Pc lan .... 2'* ..... ILitMuanwa La tvia - Kazakhstan j . - . fl * KyrgyWYz Republic|j ': . ttlungary l : :____ _ICzech R4pubIlc j _ ' ; . ;ICroatia ,Butgaria - ~~~~~~~~~~~~~~~~~~-. Btlparus - lAzerbaUan -__ __ '~~~~~~ _IAbiniall 0 2 4 6 8 10 12 14 % f GDP Source: World Bank, 2000, World Development Indicators It is difficult to determine, ex ante, the adequacy of current levels of public investments. This depends on many country-specific factors including the current capital stock, its quality, the regulatory environment and investnent climate, the availability of financing, and the development strategy of the country. However, it is clear that investment needs in infrastructure and replacement of manufacturing equipment are high, while investment levels on average are below the public investment levels in OECD countries (at over 4 percent of GDP) with much higher levels of private investment. In several industries, where physical infrastructure is important to promote the efficiency of production, marketing and distribution, security and incentives for private investment are not adequate, and public investment is needed to 'crowd in' new private investment and revitalize the industrial base. This requires a change in the 'philosophy' of public investments-away from centralized investment planning towards a partnership with the private sector, with clear criteria for state intervention, with rigorous analysis of expenditure priorities and tradeoffs. This is lacking in most countries. For instance, in Uzbekistan, almost half of public investment in the 2000 Budget went to building post-secondary 22 World Bank, 1997, Estonia Public Expenditure Review, Report No. 1 6420-EE 15 vocational colleges and lyceums with little discussion of the opportunity cost of these investments such as the economic and social costs of deferring critical investment in Uzbekistan's crumbling irrigation system on which the main export crop (cotton) and a substantial part of the population depends. B. Protecting Pro Poor Expenditures The magnitude of the cuts in the transition countries meant that the burden of the cuts had to be felt by all sectors, including social transfers.23 The dramatic nature of the income decline and attendant increase in poverty and unemployment in the transition economies in Europe and Central Asia have highlighted the challenge of providing effective social protection. Previous mechanisms of social protection which focused on those with special needs have either become irrelevant, unaffordable, or difficult to administer.24 Under central planning, the state provided full protection against income risks over an individual's lifetime through employment while of working age, and through pensions and social assistance for those retired or unable to work. With transition to the market, job security vanished, and risks of unemployment and poverty rose sharply. The transition countries have adopted two broad approaches to addressing these risks. The European transition countries-where the institutional and administrative capacity is stronger, and where economic restructuring is more advanced-let employment levels adjust and used generous social protection to lay off workers and to protect the pensioners and the disabled. These countries also adopted means-testing for social assistance payments, and began modernizing their pension systems through the introduction of multipillar systems. Real wages and real transfers have been rising in these countries. In contrast, the Eurasian countries sought to protect employment and forestall restructuring, even if that meant a sharper erosion of real wages and pensions. Pension and social assistance arrears have been pernicious, hurting their intended poor beneficiaries, and subsidies for housing and utilities have been generalized with little effort at targeting. Groups dependent on institutionalized care have often fared the worst, living in miserable conditions with dwindling public support. Informalization of labor markets has been the norm. As a result, current levels of public expenditures on social protection are almost two times higher as a share of GDP in the CEB countries than in the CIS countries (Table 2 above). A comparison of pension expenditures-the largest share of social protection expenditures-in selected transition countries with OECD expenditures suggest that most European transition economies spend as much on pensions as OECD countries though it may be high evert in some of the Eurasian countries (see Figure 9). Time series and cross country data is not available which would permit an analysis of trends in expenditures for different categories of social protection though anecdotal and country case studies provide valuable insights into the problems and the diversity of country experiences. In Georgia, for instance, aggregate expenditures cuts meant that pension payments fell to about 0.8 percent of GDP in 1994 before recovering to close to 2 percent of GDP, still a very low level. In contrast, Poland has public pension expenditures close to 14 percent of GDP. In Russia, pension expenditures in 2001 were equivalent to 5.6 percent of GDP. 23 Other transmission channels have also affected the poor adversely. For instance, the spread of informal and under-the-table payments hurt the poor, who have less capacity to pay, more than they hurt the rich. The emergence of pervasive arrears in public-sector payments, wages and pensions in parts of the CIS (e.g., Georgia, Moldova, Russia) are also highly regressive as they fall disproportionately on the poor and are highest in poorer regions. 24 See World Bank (2000b) for an evaluation of transitional experience with social protection mechanisms and strategies. 16 Figure 9: Relative Spending on Pensions: Selected Transition and OECD Countries 16.0 * haly 14.0 - * Poland * Slovena * Franro Finland * Swhaerland 12.0 Gerrny Laviva Spain * S,ed o Moldon * Hungary r 10.0 * C *nane kran UK * Dnmari Xl sh < *zch e~ 8.0 Rea * US I # Russia u *Aibanba *Canada X 4.0 - Tu ei;an *Amen1a 2.0 * Ceorgia 0.0 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 GDP per capita (1995 USS) Source: World Bank (2000b) The challenge of rationalizing social protection expenditures remains huge. For this, benefits must be adjusted to affordable levels and strictly limited to target groups, while aiming to minimize errors of inclusion and exclusion. Even in these poorer economies, targeting of cash benefits can be achieved through several instruments, including through life line tariffs, self- selection approaches, geographic and community targeting, as well as through indicators of income or income proxies. De-institutionalization in some instances, where family and community support structures are intact, may also be beneficial. Ultimately, cost efficiency measures will help but not overcome the problem of under-funding, which must be redressed through reallocation of public funds. Reform of social protection schemes often provokes political opposition. Flattening of social benefits, for example, is often seen as unfair by those who have contributed more from their incomes to the consolidated fund, as Georgia's recent experience has shown. Hence, managing the political opposition, and maintaining adequate support to politically sensitive schemes, is a constraint that many governments in the region face. Governments will need to strengthen technical capacity for actuarial forecasts of pension expenditures and revenues, auditing of pension funds, and analysis of fund collection problems in order to reduce the risks of faulty design and weak implementation. C. Reducing Fiscal Risks The reduction of fiscal risks to manageable levels is also a key dimension of the quality of fiscal adjustment. However, the trends in the transition economies are mixed. While the transition to market economy has led to the emergence of new types of fiscal risks, it has also provided th'e policymakers with new instruments for managing them. Four sources of fiscal risks are addressed here: contingent liabilities, payment arrears and budgetary offsets, institutional 17 arrangements for budget management, growing external debt service burden, and the drain from military spending. (a) Contingent Liabilities The reported fiscal adjustment does not accurately reflect actual adjustment. As recent literature on the subject demonstrates, there are many ways in which governments can comply with the quantitative targets for revenues and expenditures by switching to hidden liability accumulation or asset diminution.25 For most transition (and developing) economies, budget deficits or surpluses constitute only a portion of the total change in government liabilities, much of which derives from the realization of contingent liabilities, macroeconomic volatility, and quasi-fiscal operations of the government. These contingent liabilities arise from various sources. The largest source of these are the liabilities of state-owned enterprises and the financial system which industrial and bank restructuring and privatization programs seek to address with attendant large fiscal costs. More recent risks arise from government guarantees for enterprise borrowings, social insurance schemes (pensions, medical insurance), and local government borrowings. Pension liabilities are of particular concern as demographic trends under defined benefit schemes make current arrangements fiscally unaffordable." Contingent fiscal liabilities are very attractive since they provide a hidden source of financing, often substituting for direct government finances to sustaining economic outcomes. In the process, however, they compromise future fiscal stability by generating sudden and unanticipated claims on the budget. Indeed, results from cross country regressions suggest that the increase in average implied deficits (i.e. after including hidden liabilities) is strongly correlated with currency crises even as no such relationship is found between the conventional deficit and currency crises.27 While even OECD countries have large contingent liabilities, the vulnerability of these CEB countries to such fiscal risks is exacerbated by several factors including the prevalence of less clear ownership structures, underdeveloped regulatory frameworks, weaker enforcement of laws, shallow and nascent domestic debt markets with weak information disclosure, and limited risk management capacity. Different approaches have been made to attempt to quantify these "hidden" liabilities and deficits.28 The magnitudes of these contingent liabilities, even if only partially accounted for, are sizeable. For instance, in the Czech Republic, estimates of such hidden deficits are close to 8 percent of GDP of which explicitly and implicitly guaranteed loans accounted for about half. 29 Oftentimes, these liabilities hit the budget in a significant manner. In Bulgaria, government borrowings to help finance failing enterprises cost the government an estimated 13 percent of GDP over 1994-96. The fiscal costs of bank restructuring-some of which was prompted by banking crises-have ranged from 0.9 percent of GDP in Latvia to 22.0 percent of GDP in 25 See, for instance, Easterly (1998), Polackova (1998), or Kharas and Mishra (1999). A comprehensive assessment of fiscal adjustment needs to incorporate the net change in govermment assets and liabilities (including contingent liabilities). 26 There is a parallel with financial savings that were substantially wiped out by inflation and banking crises during the early years of transition, for which some governments may feel a liability. 27 See Kharas and Mishra (1999) 28 See, for instance, Burnside et al (1998) and Davis et al (1997) and Kharas and Mishra (1999). 29 World Bank, 1999, Czech Republic-Toward EUAccession, Washington, D.C. Bulgaria.0 In Slovakia, the stock of bonds to replace the bad loans of the banking sector will is estimated to cost around 12 percent of GDP. The growth of large public and publicly-guaranteed borrowings pose a particular problem for some of the poorest countries in the region, and constitute a serious fiscal risk. This is most pronounced for the poorer CIS countries-Armenia, Georgia, the Kyrgyz Republic, Moldova, and Tajikistan-where the total external debt burden in 2000 ranges from about 45 percent of GDP in Armenia to almost 140 percent of GDP in the Kyrgyz Republic. The net present value of future debt-service payments exceeds 150 percent of exports in Georgia, Armenia, and the Kyrgyz Republic, and do not appear sustainable.3' The CEB countries have seen smaller increases in interest costs and with stronger growth, export, and revenue performance, the economic burden has been more manageable. The rapid growth of external debt in some of the transition countries reflects several factors-including low growth, high current account deficits, and high interest rates-as well as the poor quality of fiscal adjustment policies. This has negated a positive initial condition for the CIS countries which emerged from the collapse of the Soviet Union with little external debt (as Russia took over these liabilities) and very low debt service payments. The rising debt servicing costs have reduced the fiscal space for discretionary expenditures at the same time that revenues have declined. Recognition of the problems posed by contingent liabilities, and policies to control them, have, however, shown considerable progress in some countries in recent years. This has been most notable in the CEB countries: Hungary, Estonia, and Latvia have opted for rapid sell-off of commercial banks rather than risk direct fiscal costs from rescuing them from banking crises. Hungary undertook pension reforms early which raised budget deficits temporarily but reduced long-term public pension liability. Moreover, even as the Government has taken on new fiscal risks in the form of state guarantees and growing programs of credit, it has dealt with these programs in a fiscally prudent and transparent manner.32 Across the CIS countries as well there is progress towards greater transparency in budgeting to incorporate extra-budgetary funds into official budgets, and adopt new Budget Codes incorporating fiscal management reforms (Russia, Kazakhstan). Several countries are in the midst of reforming their pension system to place them on a sustainable financial basis, including Russia which recently announced a major program of pension reform including formation of a defined contribution program. Indeed a new wave of fiscal reforms aimed at reducing fiscal risk and imparting greater transparency to the budget process appears to have gained momentum towards the end of the first decade in transition.33 (b) Quasi Fiscal Activities Even as some quasi-fiscal activities, such as off-budget funds and implicit subsidies, have been gradually incorporated into budget as an explicit fiscal measures, problems have remained. Wage and payments arrears have remained problematic in several CIS countries. Implicit subsidies to support failing enterprises and social assets devolved from pubic enterprises (housing, schools, health clinics, sanitoria) have continued in large part due to the failure of formal social safety nets and the political pressures to maintain employment in the large state sector. Even as revenue reforms have sought to reduce tax exemptions, tax expenditures remain a 30 Tang et al (2000). It is difficult to unbundle the fiscal costs from re-capitalization needs (as a result of structural reforms in banking) from government financing (to rescue banks from crises) as banking crises normally result from structural and institutional weaknesses in the banking sector. 31 See World Bank (200 lb). 32 Brixi et al (1999) 33 Martinez and Boex (2000) describe a 'third period' of fiscal management reform in transition countries in transition countries that aims to impose greater fiscal discipline and hard budget constraints on governments. 19 key element of industrial policy in many countries. Budgetary offsets developed sharply in the mid-I 990s as a way for many CIS countries to cope with the fiscal crises. Under-pricing of utilities services provides subsidies to consumers while taxing the utility providers. Although arrears and the use of non-monetary instruments have been more prevalent in Russia and other CIS countries, especially in the mid-1990s, they have declined in use and importance as the deleterious effects from these practices became evident. The problem of non- payments emerged most starkly in Russia, fueled by an inconsistent macro-micro policy mix of tight monetary targets, rapid disinflation, but weak fiscal discipline and soft budget constraints on enterprises and public utilities.34 From 1995 to 1998, the Russian Govemment sought to achieve rapid stabilization by fixing the exchange rate and tightening credit within the overall expenditure targets. But inadequate expenditure provisions (e.g. for public sector energy bills), rigidities in reallocation, and intense pressure to maintain state support in select sectors led to the use of arrears as a 'normal' financing mechanism. This led to increased public borrowings through the domestic T-bill market, increasing fiscal pressures on government from debt service. Political pressure to maintain subsidies to the largely bankrupt state enterprise sector (essentially social safety nets for these public employees), led to the use of arrears, non-cash settlements, and elaborate 'offset' procedures estimated to total 7-10 percent of GDP over 1995-97. Between 1994 and 1997 offsets increased from 11 percent to 24 percent of federal budget revenues, before declining thereafter. The web of non-payments led to chronic tax arrears, particularly from energy monopolies passing on their non-payments problem to govemment. As confidence in economic management plummeted, interest rates soared and precipitated the financial meltdown of 1998. Implicit subsidies also remain widespread, particularly in the housing and energy sector, through arrears to consumers, deferred payments, low cash collections, and utility tariffs which do not adequately compensate the utilities for the economic cost of service provision. This effectively taxes these sectors and undermines their financial soundness, acting as a disincentive to investments in the energy sector, and contributing to the deterioration of the energy infrastructure. In Azerbaijan, the quasi-fiscal activities in the energy sector (primarily from under- pricing and nonpayments) amounted to 22 percent of GDP in 1999.35 Tax expenditures-preferential tax exemptions or reductions to support specific sectors or activities-have also increased in many countries, often because they are less visible than direct subsidies and often can be passed with executive approval bypassing legislative oversight. In Poland the value of tax expenditures grew by an estimated 32 percent per year between 1992 and 1998, and were highly regressive.36 Foregone revenues from tax exemptions are pervasive. In Kyrgyzstan, it is estimated that tax exemptions cost the budget the equivalent of 5-7 percent of GDP. Arbitrary tax exemptions in Russia granted by federal and sub-national authorities are costly and seriously distort the tax base. In summary, while comprehensive data on arrears, nonpayments, tax expenditures, and implicit subsidies are hard to come by for the region, their prevalence appears to have increased in the mid-i 990s in part due to the failure of formal social safety nets in the CIS countries. As reform has deepened, efforts to tackle these problems and reduce economic distortions have increased, and their incidence has generally declined across most countries. 34 See Pinto et al (1999) for a detailed exposition of the non-payments problem and its origins, including the flawed fiscal adjustrnent. 35 Petri et al, (2001) 36 See Cavalcanti and Li (2000). 20 (c) Institutional Reforms in Budget Management The transition economies started with grossly inadequate systems of budget management which has posed a serious challenge for understanding the fiscal situation, understanding the cost and benefits of public policy measures as well as the tradeoffs implicit in public policy choices, and budgeting for results. Given the low starting point, institutional reforms in budget management, addressing both revenue and expenditure management, have been a common feature of the fiscal transition. Tax Policy and Tax Administration. The transition during the 1990s also brought about an overhaul of the tax system in many countries, with the introduction of new and more efficient instruments of taxation.37 The reform of tax policy and institutions became essential to the fiscal adjustment. The overall burden of taxation has been reduced. Tax collections saw a shift from direct taxes towards indirect taxes, and, within direct taxes, towards personal income taxes.38 At the same time, the efficiency of the tax system was enhanced as introduction of value-added taxation became a central part of the tax system in most countries, although problems with design and implementation have been persistent in some countries.39 While deficiencies in tax policy remain-for instance in the use of the origin basis for VAT taxation by Russia, the non-use of accrual basis for VAT taxation, lack of appropriate deductions from enterprise income for tax purposes, and the high payroll tax of around 35 percent for most CIS countries40-tax policy improvements have been substantial and in many countries compare favorably with modem day standards. However, even as progress has been made on tax policy issues, progress has lagged on tax administration which remains weak and fragmented in all of these countries and is the most important constraint today to improved revenue collection. This is particularly true in federated countries with a tiered tax collection structure. 4 Inter-Governmental Fiscal Relations. Public revenue and expenditure reforms through transition have fundamentally transformed the fiscal landscape of public finance between levels of government. However, problems have developed. In several countries expenditure responsibilities were devolved to sub-national governments, along with new mandates, but without fully revisiting the basis for expenditure and revenue assignments. This has led to inter- government fiscal imbalances, and has prevented the advanced reforming countries from fully exploiting the potential public finance and service delivery benefits from decentralization. In larger countries with federal structures, such as Russia and Ukraine, unclear revenue assignments and expenditure obligations in the face of a weakening state made fiscal 37 See Tanzi and Tsibouris (1999), Ebril and Havrylyshyn (1999) or Martinez-Vazquez and McNab (2000) for more on the tax reforms in transition countries. In the pre-transition economies, (i) most tax revenue was generated from three major sources-the tumover tax, the enterprise tax, and the payroll tax; (ii) taxation of resources and indirect subsidies were implicit in the state allocation of under-priced resources (relative to international prices); (iii) government payments were processed through the state's monobank; and (iv) the 'plan' formed the basis for the budget. 38 Available data suggests that on average personal income taxes rose from around 40 percent of total taxes on income and profits in 1991, to over 50 percent in 1998. Corporate taxes fell correspondingly. 39 See Summers and Sunley (1995) for a survey of early experience with VAT adoption in Russia and other FSU countries. 40 The high payroll tax is particularly onerous for-small businesses, and inhibits small business activity. 41 See Martinez and McNab (2000) for further discussion of tax administration issues for CIS countries. They note that although most CIS countries have a modem looking tax structure and broadly satisfactory tax policies, tax administration is severely deficient. 21 management more difficult. Fiscal decentralization took place early in the transition, largely motivated by the political demands for greater local autonomy though economic efficiency and macroeconomic considerations also played an important part. Thus spending responsibilities- particularly in education, health and social welfare assistance-were devolved to the subnational governments in order to relieve federal level expenditure burdens. At the same time, the federal governments maintained their revenue assignments and taxing powers. This imbalance between expenditures and revenues squeezed subnational budgets and compromised public service delivery and efficiency. Federal transfers remained inadequate to fill this gap. This created perverse incentives for discretionary fiscal behavior including bargaining not only between the federal and subnational governments, but also among the budget entities and their suppliers (such as utility companies) at the subnational level, and a systemic problem of non-payments and arrears. To the extent that subnational budgets held responsibility for social safety net expenditures, these problems had serious distributional and equity implications. There are several areas of weakness. First, although most countries have legislation that list government responsibilities at each level, detail over revenue and expenditure assignments are often lacking. Many central governments have devolved responsibilities downwards but are reluctant to give up control over revenues, or to devolve real decision making authority.42 In several of the advanced reform countries, the role of intermediate levels of government is now relatively clear, but for the third tier, which face problems of fragmented or inefficient service delivery, there is less clarity. For these countries (Czech Republic, Hungary, Latvia) exploring scale economies in service provision to improve efficiency, and organizing effective regional organizations, is a major challenge. Greater accountability of local government is essential to effective reform and achieving the efficiency goals of decentralization, but with this must come real control over budgetary resources. Local govermments need to develop their own sources of revenue (user fees, property taxes, etc.), and have greater autonomy over revenues. Second, clarifying and enforcing transparent and stable rules based transfer systems is another critical area in need of further reform. Elaborate transfer formulas can be undermined through end-year negotiated transfers that also serve to soften budget constraints and direct attention towards political negotiations rather than efficient fiscal management, as in Bulgaria (and several CIS countries). Third, linked to this is the need to oversee and monitor local govermment borrowing, which has grown rapidly in recent years as an autonomous source of local financing. Effective regulation from the center and rigorous standards of local government accountability are required. Initially limiting borrowings to investment financing (the 'golden rule') where local government fiscal responsibilities are less well developed, should be considered. Strong information reporting is needed, as well as building greater capacity in sub-national govemments in public expenditure and debt management. Budget and Expenditure Management. Improvements in resource allocation need to be accompanied by effective tools for budget monitoring, execution, and cash management. Most of the less advanced market reformers have embarked on developing modem treasury systems, with the aim of providing comprehensive payment, accounting, and financial reporting services to the central govemment.43 This requires fundamental reform of existing institutions and processes which has lagged in the transition economies. Progress with treasury reforms has been slow, and in some countries, including Russia, their scope in still much too limited. While no treasury in any of these countries is fully functional, progress in Kazakhstan is the most advanced. In most 42 These issues are explored in greater detail in World Bank (200 la). 43 See Potter and Diamond (2000). 22 of these countries, much better progress has been made in setting up the treasury payment system and a basic treasury single account than in introducing a general ledger system for government accounting or developing capacity for financial management. Systemic arrangements for commitment control remains elusive. While the development of treasury systems will bring gains from better cash management, the achievement of the goals of good public expenditure management will be elusive if wider issues of budget management and poor governance are not addressed. Indeed, insufficient provisions for some public expenditures-especially energy-risk leading to the development of large payments arrears and thereby undermining macroeconomic management, as painfully evidenced, for example, in Russia during 1996-99. Processes of budget preparation need to be strengthened to ensure that (i) budgets are realistic, (ii) expenditure and revenue planning is rooted in a macroeconomic framework, (iii) resource allocation decisions reflect the stated policy priorities of the government, (iv) budget classification systems permit a program orientation with clarity towards the unit of appropriation, and (v) adequate systems of post-budget evaluation and audit are developed. Transparency in budget preparation, evaluation, and audit are particularly important to develop greater accountability over the use of public resources. Financing Instruments. During this process of fiscal adjustment, a shortage of financing instruments have both served to limit the size of the fiscal balances-a salutary contribution- while making macroeconomic management more difficult. First, privatization revenues fell far short of expectations and did not generate the budgetary financing (or expenditure savings) that were anticipated.44 The low returns reflected the widespread use of vouchers, asset stripping, weak foreign participation (due to the uncertain investment climate), and a general shortage of buyers for unprofitable enterprises.45 Second, foreign financing of the budget has been limited except for some of the poorer countries, such as Georgia and the Kyrgyz Republic. Despite the upsurge in sovereign borrowings in private capital markets in the mid-1990s, most foreign financing has been from official sources. Program-based disbursements of foreign aid were significant during the early years but are now less so. Third, domestic bank financing emerged as the primary source of budgetary financing. Finally, arrears and non-monetary instruments became a major source of de-facto financing as government and enterprises tried to forestall massive unemployment in bankrupt public enterprises, as elaborated below. IV. CONCLUSIONS AND POLICY LESSONS Ten years into transition, a wide variety of fiscal outcomes characterize the transition countries. Notwithstanding the dramatic fiscal adjustment undertaken by these countries in quantitative terms, the imperative of macro-stabilization and the weakness of underlying institutional capacity served to compromise the quality of fiscal adjustment, particularly in the CIS countries. Four broad conclusions can be drawn. First, investments in public goods-public infrastructure, health and education services- fell in both absolute and relative terms. The impact of this would have been minimized if there were commensurate increases in private sector investments or improved efficiency in public service delivery. But with limited private sector capacity and a weak legal and institutional environment for private providers, private sector response has been marginal. At the same time, public sector capacity and incentives for improved performance remain weak. Given the under- 44 See Tanzi and Tsibouris (1999) on magnitudes of privatization proceeds across the transition econornies. 45 See Cheasty and Davis (1996) for a summary discussion of these emerging trends in mnid-decade. For a discussion of recent trends, see Tanzi and Tsibouris (1999). 23 developed budget and expenditure management practices, the sharp fiscal contraction led to declining health, and education outcomes in the region. Second, reduced spending on govemment transfers contributed to the sharp deterioration in income inequality in the CIS countries.46 Some deterioration in income inequality is arguably inevitable in the move to market based system. However the absence of adequate social safety nets (dysfunctional unemployment insurance, and social welfare assistance) and the decline in public spending, which hit pensions and family allowances sharply, disproportionately hurt the poor. At the same time, studies have shown that sustaining transfers in the CEB countries, and improvements in targeting, reduced the average rise in inequality. Third, fiscal risks have likely increased during transition as the debt burden increased in most countries, and as exposure to global commodity and financial markets caused greater volatility in commodity prices and risks of financial contagion (as amply demonstrated by the East Asian and Russian financial crises of 1997-98). While financial sector and other structural reforms provide some mitigation of these types of risk, additional fiscal risks have emerged from the costs of social and structural reforms, particularly in the banking sector, in pensions, and in enterprise restructuring. Fourth, high expenditures in the CEB may have been critical to maintaining political and popular support for reforms, and hence contributed to the recovery of growth in the first decade of transition. Expenditures in CEB countries were maintained at high levels, especially in the social sectors, which helped cushion the impact of transition and made reforms more acceptable.47 By contrast, the erosion of social transfers and public provision of social services has impacted severely on the population in many of the CIS countries and served to sustain more entrenched opposition to the reform process. However, at the start of the second decade of transition, these expenditures in CEB countries may be unsustainably high in some countries, and may now act as a drag on growth. Now that macroeconomic stabilization has been largely achieved, even if still fragile in many CIS countries, policy makers need to tum their attention to improving the quality of the fiscal adjustment. In broad terms, a differentiated agenda for fiscal reforms emerges. For the more advanced CEB countries, the pressing agenda is to move forward with the second generation of fiscal reforms: reducing fiscal risks, improving the quality of intergovernmental finances, and reducing the size of the state. For the less advanced market reformers, mainly in the CIS, policy makers need to turn their attention to more fundamental issues of fiscal sustainability and use of the budget as an effective instrument to meet the social and policy needs. Priorities are strengthening basic fiscal discipline, improving the efficiency of resource allocation and use, and strengthening the institutional arrangements for better fiscal management. Indeed, for most of the transition countries, it is only by strengthening the incentives embedded in these institutional arrangements that policies, resources, and outcomes be aligned, and fiscal policy be made more effective. 46 See World Bank (2000). Between 1987-90 and 1996-99, the average Gini measure of income inequality increased for CIS countries by 0. 15, and by less than half this amount-0.07-in the CEB countries. While an increase in income inequality during transition was anticipated and not necessarily a bad development-reflecting rising returns to education, the decompression of wages, and returns to risk- taking and entrepreneurship- the extent of the increase in some countries has raised worries of social tensions, and also reflect lack of opportunities, weak capabilities, and increasing economic insecurity for groups within the population. 47 Research on Poland (Keane and Prasad (1999)), for example, shows that pensions were critical to preventing the elderly from falling into poverty whereas the erosion of pensions in Russia had a severe impact on many elderly. Details of pension erosion in Russia are in World Bank (1996). 24 References Brixi, Ghanem, and Islam, 1999, "Fiscal Adjustment and Contingent Government Liabilities", Policy Research Working Paper Series No. 2177, World Bank. Brixi, Papp, Schick, 1999, "Fiscal Risks and the Quality of Fiscal Adjustment in Hungary", Policy Research Working Paper Series No. 2177, World Bank. Burnside, C., M. Eichenbaum, and S. Rebelo, 1998, "Prospective Deficits and the East Asian Crises", Policy Research Working Paper Series No. 2174, World Bank. Cavalcanti, Carlos and Zhicheng Li, (2000), "Reforming Tax Expenditure Programs in Poland", Policy Research Working Paper Series No. 2465, World Bank. Cheasty. A. and Jeffrey Davis, 1996, Fiscal Transition in the Countries of the Former Soviet Union, Working Paper No. 96/61, IMF. Davis, D., A. Daniel, and A.M. Wolfe, 1997, "Fiscal Accounting of Bank Restructuring", Working Paper Series No. PPAA/97/5, IMF, De Melo, Denizer, Gelb, and Tenev, 2001, "Circumstance and Choice: The Role of Initial Conditions and Policies in Transition Economies", World Bank Economic Review, Vol. 15. No. 1. Easterly, W., 1998, "When is Fiscal Adjustment an Illusion?", Policy Research Working Paper No. 2109, World Bank, EBRD, 1999 Transition Report. Ebrill, L. and 0. Havrylyshyn, 1999, "Tax Reform in the Baltics, Russia, and Other Countries of the Former Soviet Union", Occasional Paper No. 182, IMF. Fischer and Sahay, 2000, "The Transition Economies after Ten Years", Working Paper No. WP/00/30, IMF. Gray, Dale F., 1998, "Evaluation of Taxes and Revenues from the Energy Sector in the Baltics, Russia, and Other Former Soviet Union Countries", Working Paper No. WP/98/34, IMF. Gupta, S., Luc Leruth, Luiz de Mello, and S. Chakravarti, 2001, "Transition Economies: How Appropriate is the Size and Scope of Government?", Working Paper Series WP/01/55, IMF. Hagen, J. and Ian Harden, 1996, "Budget processes and Commitment to Fiscal Discipline", Working Paper Series No. 96/7 8, IMF. Kharas, H. and Deepak Mishra, 1999, "Hidden Deficits and Currency Crises", mimeo, World Bank. Klugman, J., 1999, "Financing and Governance of Education in Central Asia" MOST: Economic Policy and Transitional Economies, No. 9:423-42. Lazear, Edward P. (editor), 1995, Economic Transition in Eastern Europe and Russia: Realities of Reform, Hoover Institution Press, Stanford University. 25 Martinez-Vazquez, Jorge, and Robert McNab, 2000, "The Tax Reform Experiment in Transitional Economies", Working Paper 00-6, Georgia State University International Studies Program. Martinez-Vazquez, Jorge, and Jameson Boex, 2000, "Budgeting and Fiscal Management in Transtional Economies", Working Paper 00-1, Georgia State University International Studies Program Petri, Martin, Gunther Taube, Aleh Tsyvinski, 2001, "Energy Quasi Fiscal Activities in BRO Countries", mimeo, IMF. Pinto, B., V. Drebentsov and A. Morozov, 2000, "Dismantling Russia's Nonpayments System", Technical Paper Series, No. 471, World Bank. Polackova, H., 1998, "Contingent Liabilities: A Hidden Risk for Fiscal Stability", Policy Research Working Paper Series No. 1989, World Bank. Potter, B. and J. Diamond, 2000, "Setting Up Treasuries in Former Soviet Union", Occasional Paper Series No. 198, IMF Stiglitz, J., 1999, "Whither Reforms?", Paper presented at the World Bank Annual Conference on Development Economics, April. Summers, Victoria P. and Emil M. Sunley, 1995, "An Analysis of Value-Added Taxes in Russia and Other Countries of the Former Soviet Union", Tax Notes International, No. 10:2049-72, June 19. Sundberg, Mark and Alexander Morozov, 2000, "Benchmarking Public Expenditure in the Russian Federation: Mystery, Mismeasurement, and Mismanagement", in Proceedings of the 92 d Annual Conference of the National Tax Association, Atlanta, Georgia. Tang, H, Edda Zoli, and Irina Klytchnikova, 2000, "Resolving Banking Crises in Transition Countries: Fiscal Costs and Related Issues", mimeo, World Bank. Tanzi, V. and Tsibouris, 1999, "Fiscal Reform over Ten Years of Transition", Paper presented to the Fifth Dubrovnik Conference on Transition Economies, July 23-25, IMF. Wallich, Christine (ed), 1994, Russia and the Challenge of Fiscal Federalism, World Bank Regional and Sectoral Studies, The World Bank, Washington D.C. World Bank, 2002, Transition: The First Ten Years, Washington D.C. World Bank, 2001 a, Decentralization in the Transition Economies: Challenges and the Road Ahead, Poverty Reduction and Economic Management Unit, Europe and Central Asia, Washington D.C. World Bank, 200lb, Armenia, Georgia, Kyrgyz Republic, Moldova, and Tajikistan: External Debt and Fiscal Sustainability, Board Paper, Washington D.C.. World Bank, 2000, Making Transition Work for Everyone, Washington, D.C. 26 World Bank, 2000b, Balancing Protection and Opportunity: A Strategy for Social Protection in Transition Economies, Washington, D.C. World Bank, 1999, "Russia: Regional Education Study", Europe and Central Asia Region, Human Development Sector Unit, Washington, D.C. 27 Policy Research Working Paper Series Contact Title Author Date for paper WPS2806 Dirty Exports and Environmental John S. Wilson March 2002 P. Flewitt Regulation: Do Standards Matter to Tsunehiro Otsuki 32724 Trade? WPS 2807 The Role of Natural Resources in Benoit Bosquet March 2002 D. Duff Fundamental Tax Reform in the 39506 Russian Federation WPS2808 A Capital Accord for Emerging Andrew Powell March 2002 E. Mekhova Economies 85984 WPS2809 On the Measurement and Impact of Robert D. Ebel March 2002 M. Morris Fiscal Decentralization Serdar Yilmaz 37285 WPS2810 Growth, Distribution, and Poverty Luc Christiaensen March 2002 N. Nouviale in Africa: Messages from the 1990s Lionel Demery 34514 Stefano Paternostro WPS2811 The Epidemiological Impact of an John Stover March 2002 H. Sladovich HIV/AIDS Vaccine in Developing Geoff P. Garnett 37698 Countries Steve Seitz Steven Forsythe WPS2812 Can Financial Markets be Tapped to Jerry Skees March 2002 P. Kokila Help Poor People Cope with Weather Panos Varangis 33716 Risks? Donald Larson Paul Siegel WPS2813 The Collective Model of the Household Kaushik Basu March 2002 N. Jameson and an Unexpected Implication for Ranjan Ray 30677 Child Labor: Hypothesis and an Empirical Test WPS2814 Estimating the Endogenously Gayatri Koolwal March 2002 N. Jameson Determined Intrahousehold Balance Ranjan Ray 30677 of Power and Its Impact on Expenditure Pattern: Evidence from Nepal WPS2815 Pricing Currency Risk: Facts and Sergio L. Schmukler March 2002 E. Khine Puzzles from Currency Boards Luis Serv6n 37471 WPS2816 Explaining the Migration of Stocks Stijn Claessens March 2002 E. Khine from Exchanges in Emerging Daniela Klingebiel 37471 Economies to International Centers Sergio L. Schmukler WPS2817 Does Sequencing Matter? Regulation Scott Wallsten April 2002 P. Sintim-Aboagye and Privatization in Telecommunications 37644 Reforms WPS2818 Corporate Governance, Investor Leora F. Klapper April 2002 A. Yaptenco Protection, and Performance in Inessa Love 31823 Emerging Markets WPS2819 Goals for Development: History, Shantayanan Devarajan April 2002 S. Brickland Prospects, and Costs Margaret J. Miller 30944 Eric V. Swanson Policy tResearch Working Paper Series Contact Title Author Date for paper WPS2820 The Privatization of the Russian Igor Artemiev April 2002 V. Joseph Coal Industry: Policies and Processes Michael Haney 32155 in the Transformation of a Major Industry WPS2821 Income, Wealth, and Socialization Daniel Lederman April 2002 P. Soto in Argentina: Provocative Responses 37892 from Individuals WPS2822 An Econometric Analysis of the David Mckenzie April 2002 C. Mendoza International Bank for Reconstruction 80599 and Development's Creditworthiness WPS2823 Real Exchange Rate Uncertainty Luis Serven April 2002 P. Soto and Private Investment in Developing 37892 Countries WPS2824 Trade Policy and Labor Services: Maurice Schiff April 2002 P. Flewitt Final Status Options for the West 32724 Bank and Gaza WPS2825 Demand for Imports in Venezuela: Mario A. Cuevas April 2002 M. Geller A Structural Time Series Approach 85155 WPS2826 Potential GDP Growth in Venezuela: Mario A. Cuevas April 2002 M. Geller A Structural Time Series Approach 85155 WPS2827 Learning to Export: Evidence from Marcel Fafchamps April 2002 E. Khine Moroccan Manufacturing Said El Hamine 37471 Albert Zeufack WPS2828 Beyond Oaxaca-Blinder: Accounting Francois Bourguignon April 2002 F. Bourguignon for Differences in Household Income Francisco H. G. Ferreira 31056 Distributions across Countries Phillippe G. Leite WPS2829 The Spatial Distribution of Poverty Nicholas Minot April 2002 R. Bonfield in Vietnam and the Potential for Bob Baulch 31248 Targeting WPS2830 Labor Market Implications of Lodovico Pizzati April 2002 L. Pizzati Switching the Currency Peg in a 32259 General Equilibrium Model for Lithuania WPS2831 Health Outcomes in Poor Countries Limin Wang April 2002 L. Wang and Policy Options: Empirical Findings 37596 from Demographic and Health Surveys WPS2832 Poverty and Survival Prospects Adam Wagstaff April 2002 H. Sladovich of Vietnamese Children under Doi Moi Nga Nguyet Nguyen 37698 WPS2833 School Choice, Student Performance, Emiliana Vegas April 2002 H. Sladovich and Teacher and School Characteristics: 37698 The Chilean Case WPS2834 Investor Protection, Ownership, and Charles P. Himmelberg April 2002 K. Labrie the Cost of Capital R. Glenn Hubbard 31001 Inessa Love