67932 Russia: Issues in Public Expenditure Policy by Alexander Morozov, ECSPE, The World Bank Mark W. Sundberg, SASPR, The World Bank* Conference on Post-Election Strategy Moscow, April 5-7, 2000 Introduction 1. This note briefly outlines what we see as key public expenditure policy issues facing Russia today.1 Dramatic changes have taken place on the public expenditure landscape since the start of transition. In the first year after the dissolution of the Soviet Union, total recorded public expenditures captured over two-thirds of total GDP. Within the next two years this was reduced to 46% of GDP, and by 1998 consolidated budget and extra-budgetary recorded expenditures reached an estimated 39% of GDP. These figures are summarized in Table 1. Table 1: Russian Total Public Expenditure by Level of Government (1992-1998) % of GDP Federal Federal Fed. Gov’t Regional Consolidated Extra- Enlarged Government 1 transfers (net of Government Budget Budgetary Budget 3 transfers) Funds 3 1992 25.2% 1.8% 23.4% 13.0% 36.1% 33.0% 69.1% 1993 19.9% 2.7% 17.2% 17.0% 34.1% 15.1% 49.2% 1994 23.0% 3.5% 19.5% 17.5% 36.9% 9.1% 46.0% 1995 18.1% 1.8% 16.3% 14.5% 31.5% 9.5% 40.9% 1996 20.2% 2.3% 17.8% 15.6% 33.4% 10.0% 43.4% 1997 20.1% 3.2% 16.9% 18.1% 35.0% 11.1% 46.1% 1998 14.5% 1.7% 12.8% 15.2% 28.0% 11.0% * 39.0% Source: Fiscal Management in Russia. The World Bank, 1996; Russia Ministry of Finance, Staff estimates 1 Federal Government expenditures include intergovernmental transfers, which are netted out from the consolidated budget 2 Combined Consolidated Government and Extra-Budgetary Funds. Reported on a cash basis. 3 Until 1994 debt service payments on hard currency debt of the government were recorded under EBF financing. * Jan. – Sept. 1998 actual, Q4 1998 EBF accounts estimated. 2. This overall reduction of recorded public expenditure in relation to GDP has been accompanied by fairly substantial shifts in the structure of spending. Extrabudgetary funds, which accounted in 1992 for almost half of total spending, today account for about one third – largely as a result of an elimination of off-budget import subsidies and * The findings, interpretations, and conclusions expressed in this paper are entirely those of the author and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this paper and accepts no responsibility for any consequence of their use. 1 The note does not deal with intergovernmental fiscal issues, which are treated in a companion note. incorporation in the budget of remaining hard currency operations; federal spending, which in 1992 accounted for over 65 percent of consolidated federal-regional government spending, today accounts for less than one-third of the total; expenditures for National Economy (which primarily reflect subsidy outlays), declined from 11.4 percent of GDP in 1992 to roughly 6 percent, with the federal share declining from more than 50 percent to less than 15 percent; reported defense expenditures declined from about 5 percent of GDP to 2-3 percent range. 2 Accurate Measurement of Public Expenditures 3. The quality and coverage of public expenditure data currently reported by the GoRF are inadequate for public expenditure policy analysis. Ideally, the budgeted and accounted public expenditures thereunder would present a comprehensive and accurate picture of the government’s use of resources. Few if any budget or reporting systems actually employed today fully approximate this ideal. In Russia, however, there are reasons to be concerned that currently reported expenditures seriously understate the extent of what is, in effect, tantamount to public expenditure, and also result in a distorted picture of the sectoral and functional distribution of expenditure. 4. Although budget data have improved, there are many significant off budget expenditure items whose magnitude is not reflected in the budget accounting of public expenditures. Formally, they are made either in the form of off-budget accounts of budgetary entities, opened independently from regular budget accounts, or in the form of extra-budgetary funds, regulated by special legal acts. In the latter case, funds may be a separate legal entity or they may subordinate to, and be administered by, budgetary entities. An inventory of balances of off-budgetary accounts performed in 1998 revealed that they amounted to Rb 15 bn (0.6% of GDP). 3 This compares to the Pension Fund which maintained balances of less then Rb 3 bn and annual spending of around 7% of GDP). The level of total expenditures from these off budget accounts is unknown, but are likely to be at a minimum twice the level of account balances (1.2% of GDP) and may exceed 10% of GDP. 4 5. In addition to off-budget accounts of budgetary entities, which are formally legalized and regulated, it is a common practice that the personnel of these entities are often paid a fee in cash ‘under the table’ for access to formally ‘free-of-charge’ public services and/or for better quality services. Such money flows are not registered and accounted for anywhere, constituting de facto a kind of public sector informal activity. 6. At the regional and municipal levels the number of existing extra-budgetary funds is much bigger than at the federal level. In many regions the large Regional Road Funds are not consolidated into the regional budget. Empirical evidence suggests that richer regions have more extra-budgetary funds and higher proportions of off-budget to budget operations. Moscow, Yakutia, and Komi are all wealthier regions and known for their 2 Defense expenditures reported by the GoRF differ – in some cases quite significantly -- from some estimates of Russia’s defense spending reported by other sources. 3 Figure publicly announced by the Treasury, MoF. 4 The Treasury has recently started collecting information from line ministries on off-budget accounts. However, no reporting has been made available yet. 2 large extra-budgetary funds. Moscow alone has three big investment funds off-budget that have been used to finance large infrastructure projects with commercial application, among other activities. In addition to this, each Moscow prefecture has its own extra- budgetary fund.5 With effectiveness of the Budget Code beginning in 2000, these extra- budgetary funds are supposed to be consolidated in their respective budgets, although it is not yet clear how completely the authorities are complying. 7. In addition to these flows, nonpayments in tax and expenditure transactions involving the government generate additional implicit expenditures that are frequently of a quasi-fiscal nature. One recent estimate places implicit subsidies provided to enterprises through tax arrears and offsets at over 10 percent of GDP (see Table 2), and finds that, while explicit budget subsidies have declined, the sum of implicit and explicit subsidies may actually have increased.6 8. The exact magnitude notwithstanding, these are all important sources of quasi- public spending which is not including in any measurement of public finance in the economy. It is thus likely that the real weight of fiscal activity in Russia is unknown but likely much higher than the official enlarged government expenditure figure of 39% of GDP in 1998 shown in Table 1 above. The ability to conduct adequate public expenditure policy analysis depends crucially on the development of an accurate, comprehensive, consistent accounting of the government’s fiscal operations. This should be a very high priority. Table 2: Budget subsidies to enterprise sector, 1994-1998 1994 1995 1996 1997 1998 Explicit budget subsidies to enterprise sector/ GDP (percent) 10.2 8.6 7.9 8.6 5.9 Implicit budget subsidies to enterprise 0.7b/ 3.1b/ 7.6 7.4 10.4 sector/ GDP (percent) a/ Total budget subsidies to enterprise 10.9 b/ 11.7 b/ 15.5 16.0 16.3 sector/ GDP (percent) a/ Does not include subsidies in form of ad hoc tax exemptions as well as net increase in overdue fines and penalties b/ Does not include subsidies implicit through regional offsets Source: Staff estimates. Size of Government 9. Based on official data, overall spending, while declining dramatically as a share of GDP since the beginning of the transition, at about 40 percent of GDP remains very high at about 10 percentage points of GDP higher than in non-transition countries at 5 In Samara region local extrabudgetary funds are established even in rural settlements, comprised from entertainment fees, sales from articles made by secondary students, and voluntary contributions. These may be appropriately kept off budget and used for unbudgeted special purposes. 6 Recent information indicates a sharp decrease in offset operations at federal and subnational budgetary levels in 1999 and 2000. This is to be greatly lauded if it proves sustainable. 3 comparable levels of per capita GDP. 7 Put otherwise, non-transition countries with expenditure ratios of the order of 40 percent of GDP in the mid-1990s typically had dollar per capita incomes about 2.6 times larger than that of present day Russia (i.e., US $ 6,075 versus the present US $ 2,320 in Russia). 8 This, together with the fact that Russia has had great difficulty in raising its government revenue ratio to 40 percent of GDP and evidence that there is considerable inefficiency remaining in spending suggest that a substantial fraction of future fiscal adjustment may perhaps best come from reductions on the expenditure side of the ledger. However, determining the optimal size of government in Russia requires more clarity on the role of government, the sustainabilty of public finances given Russia’s institutional capacity and the impact of government on retarding private growth, the scope for efficiency gains through outsourcing of public services, etc.. 10. Any such reductions would, of course, need to be phased-in in light of both social and political considerations and with a careful assessment not only of near-term pros and cons of alternative expenditure levels, but also longer term appropriateness of alternative stances with respect to the role of the public sector and consequent overall levels of public expenditure. For example, if Russia were growing rapidly and/or pursuing policies that were likely to put the economy on a rapid growth path, one might argue that the economy could “grow into� the present expenditure ratio and that it may be optimal to maintain present expenditure levels or, perhaps, moderate cuts in this ratio. It could, however, take a very long time to grow into Russia’s present expenditure ratio. If the Russian economy were growing at 5% per annum in per capita terms, for example, it would take nearly 20 years for Russia’s per capita income levels to reach a level consonant with current expenditure percentages in non-transition economies. 11. It might also be argued that Russia and other transition economies could elect to be "different" and to sustain higher levels of transfers, public investment and consumption than cross country income expenditure comparisons with non-transition economies would otherwise indicate. There is, to be sure, some scope for such a choice, particularly during the transition period when such expenditures – efficiently deployed – can be a potent force for restructuring the economy. In the longer run, however, this scope is bounded by the constraints that political and economic freedom in an increasingly integrated global economy place on any government. Attempts to maintain too large a public sector could encourage flight of both capital and labor, or otherwise impose high efficiency costs on the economy and thus reduce growth.9 Moreover, expenditure levels are not entirely exogenous. To a large extent they are determined by the availability of revenue resources, sustainability of the budget deficit, and public debt. Public expenditures must be considered in the context of overall macroeconomic policy. 7 This estimate probably understates the size of the public sector in Russia’s economy relative to those in other countries for several reasons, as noted above. 8 The per capita GNP used in the cross country analysis described above is determined by the Atlas Method and is for 1998. If the recent real depreciation of the ruble is maintained, the analysis reported here would imply the need for still further reduction in the expenditure ratio to achieve comparability with comparator countries. 9 Cross country growth studies (e.g., empirical results reported in Economic Growth, Robert J. Barro and Xavier Sala-I-Martin, McGraw Hill) generally find that one opportunity cost of excessively large government sectors is, other things being equal, lower growth. 4 12. Notwithstanding significant changes during the last 8 years in the size and organization of government, a comprehensive public re-examination of the roles, functions, organization and future cost of government has yet to be undertaken. The question of size/overall level of expenditure can only be considered in conjunction with a forward looking reconsideration of the functions of government, the benefits associated with those functions in relation to the feasibility and opportunity cost to the economy of raising the resources to finance them. This is, we believe, a critical issue at this juncture.10 13. This re-examination of government’s role and functions would necessarily have to take into consideration the increasing role that the private sector has started to play in providing services that were formerly 100% provided by the state. Devolving financing and/or service provision responsibilities to the private sector should be considered as a serious alternative to keeping them in the state hands from efficiency point of view. Similarly, consideration of the quality of services would be one of the major parameters of this re-examination.11 These are the sort of questions which are addressed in the context of sectoral public expenditure reviews (PER). Planned And Executed Budget Outcomes 14. One indicator of the efficiency of the budget planning and execution process is the similarity of budget outcome to budget plan. In an environment of major macroeconomic uncertainty, this is rendered more difficult, and preserving budget flexibility to deviate from plan may be desirable. Greater budgetary risk would recommend pursuing a more conservative budgetary approach, with more cautious macroeconomic forecasts. However, in Russia growth parameters have consistently been overestimated, which has in part explained systematic overestimation of revenues in the budget, and a consequent need for repeated expenditure sequestration.12 15. Partly as a consequence of this, Russia has run unsustainable structural fiscal deficits since transition began, and has targeted fiscal correction as a key element of macro stabilization efforts under IMF supported programs. Accurate budget planning and execution is central to effective fiscal policies, and depend as well on the institutional capacity of government to effectively control and monitor spending through the federal treasury system. Deviations from budget plan also occur through reallocation across budget recipients within overall spending limits, which can also undermine or even vitiate public policy. 16. Several points emerge from this analysis. First, the aggregate revenue and expenditure figures indicate there has been little or no progress in moving towards 10 A related paper on Public Administration and Civil Service Reform further outlines the issues that would need to be addressed in such an undertaking. 11 Numerous data suggest significant deterioration in quality of various core services provided by the state (e.g. health, education, law enforcement, and defense) reflecting falling effectiveness of the use of state resources. 12 Macro parameters have been highly politicized in the budget process, and contributed to undermining realism and objectivity of the budget. In four of five years 1994-1998 GDP growth was overestimated in the budget. 5 bringing actual outcomes closer to planned outcomes despite achieving greater macro stability through 1997. Second, revenue shortfalls have forced parallel expenditure cuts to avoid failure in meeting deficit targets (and monetization of the deficit, in violation of IMF conditions). But expenditure cuts have generally been less than revenue cuts (by an average 6%), particularly in the early years of transition, with consequent widening of the overall deficit. 1998 was exceptional, since the government defaulted on interest payments, and forfeited access to credit markets. This analysis suggests serious questions about the overall effectiveness of budget planning and execution. 17. There have been significant deviations of actual execution from the plan across expenditure and revenue categories, even in years when total revenues and expenditures have been relatively close to plan. In 1995 when both expenditures and revenues were within 5 percent of plan, the weighted index shows that there were far larger deviations across expenditure categories. This suggests that within the budget year, expenditure reallocation (or tax compliance rates) have deviated significantly from plan or expectations. There is also evidence to suggest that deviation from across-the-board sequestration has been large. In 1993 and 1996 expenditures deviated from the average revenue shortfall by over 20% showing very large reallocation of expenditures within the budgetary year have taken place. 18. In summary, this analysis suggests the following: • Federal government revenues and expenditures have deviated significantly in actual execution from the budget plan. There is scant evidence that budget planning and projections are improving, as judged by this measure. • Expenditure cuts have deviated significantly in actual budget execution from the proportional appropriations in the budget plan. In other words, executed budgets not only show adjustment in spending levels, but major reallocation among expenditure categories from original budget plans. Sequestration has not been proportional. • The pattern of reallocation suggests revealed preferences of policy makers mid- year, or of the spending authority of different budget organs. National economy outlays on construction, investments, agriculture, and health care have disproportionately suffered, whereas law enforcement, government pensions, and intergovernmental transfers have been relatively protected from sequestration. • In this environment, budget outcomes may fail to reflect policies priorities and the value of expenditure planning can be highly compromised Subsidies, Social Security and Welfare 19. In broad terms, two categories of expenditures account for Russia’s excess (in relation to its per capita income) expenditures as per official accounts: (i) expenditures on social security and welfare (9 percent of GDP in Russia versus approximately 4 percent of GDP in Low-Middle Income Countries and 5.3 percent in Upper-Middle Income 6 Countries); and (ii) housing and community amenities (3.8 percent of GDP in Russia versus 0.8 percent of GDP in Low-Middle Income Countries and 1.5 percent of GDP in Upper-Middle Income Countries). 20. Social Security and Welfare. Expenditures on social security and welfare not withstanding their significant decline in relation to GDP are still two to three times those of comparable income non-transition comparators. Besides budgeted social expenditures, there are extra-budgetary expenditures under four major EBFs at the federal level, each with a specific social insurance function: the Pension Fund (PF), Social Insurance Fund (SIF), Fund for Obligatory Medical Insurance (FOMI), and the Federal Employment Fund (FEF). 21. There is a clear rationale for public sector participation in each of these areas. The major near-to-medium term issues with respect to these funds have more to do with their effectiveness and efficiency in providing social protection and facilitating the smooth functioning of markets than overall level issues.13 Indeed, it may be the case that transitional spending in these areas may need to increase, financed by cutting explicit and implicit production subsidy outlays. 22. There are three effectivenesss/efficiency issues that merit attention:14 (i) the scope and nature of public sector involvement and resulting size (affordability) of spending levels in the areas covered by the EBFs vis-a-vis the social benefits achieved; (ii) financing; and (iii) modus operandi of these funds.15 Efficient specific institutional and financing forms of this participation are unlikely coincide with the existing arrangements. Provision of some social benefits (unemployment benefits and retraining) can be better done from the regular budget, and provision of others (sick benefits, a part of pension benefits) – devolved to the private sector. 23. Efficiency and effectiveness issues with respect to the social insurance funds are treated in a companion note on Issues in Social Protection. Financing considerations reinforce these efficiency and effectiveness concerns: All of these funds financed through payroll taxes, totaling 38.5% of the wage bill. 16 Consequently they have a potentially major impact on employment, formal-informal composition of activity, aggregate savings, household incomes, and tax incidence in the economy. The burden of the social EBFs on firms and labor has been a recurrent theme of concern by the private sector. These funds also represent a significant contingent liability facing the state which is not explicitly considered in annual budgets. 24. With respect to modus operandi, the financial management controls and procedures that apply to these funds do not meet minimum standards for fiduciary management of public funds. 13 With the exception of the SIF, which finances a good deal of expenditures not related to social insurance. 14 [Cross reference to treatment of these issues in Social Protection note]. 15 Reforms in these areas are treated in the Social Protection note 16 Proposals introduced to the Duma in 1998 would have reduced contributions to the PF and SIF in 1999, reducing the total payroll tax burden to around 31% of the wage bill, mainly via reduced contributions to the PF. These were rejected but are actively debated as the tax burden on labor is considered too high. 7 25. Subsidies. Explicit subsidies, particularly those for utilities and communal services – which are largely untargeted, are also well above comparator norms. Considerable scope exists for reducing these expenditures (perhaps, in part, to finance increases in others related to social protection) without significant social impact. Defense 26. As noted above, Russian defense expenditures reported in the federal budget showed total outlays of 2% to 3% of GDP between 1995 and 1998. This places Russian defense spending in the bottom half of countries internationally. 17 Nonetheless, they remain about 20- 30 percent of federal non-interest expenditures and thus merit close scrutiny in any analysis of expenditure policies. 27. This is not possible at present. The Russian defense establishment operates as an enclave that is very independent of the rest of the budget and budgetary practice. The defense establishment operates with a separate budget classification system and budgetary jurisdictions than the rest of government, making it impossible to integrate defense expenditures with the rest of the budget. For this reason, the Federal Treasury is unable to monitor or manage any defense expenditures through the treasury system which is mandatory for all other ministries and agencies to use. The defense budget is also classified and kept on an accounting system that is entirely separate from the rest of the budget, again preventing integration of budgetary accounts.18 28. Irrespective of decisions that may be made for national security reasons with regard to the presentation of data, effective budgeting and control will only be possible with full comparable accounting, consistent with treatment of the rest of the public sector, of the resources devoted to defense. Pilot programs have been launched to identify the obstacles to integrating the MoD with the rest of the budget, and since end-1999 the head office of the MoD has been moved into the Treasury system, with plans to fully integrate the second and third-tier expenditure branches by mid-2000. These reforms recognize the importance of both adequate efficiency of expenditure management and preserving necessary secrecy standards for the military. Civil Service Reform 29. Civil service reform issues are intimately related to issues concerning the size and function of government. As of January, 1998 there were 6.5 million people fully or partially employed through federal budget funds (equivalent to about 6.5% of the labor force or 4.4% of the population).19 It is difficult to reconcile this number with available 17 As noted, alternative estimates place defense spending at a considerably higher figure, ranging between 5 and 10 percent of GDP. In part this is due to classification of explicit budget expenditures. In part it is also due to treatment of implicit subsidies. In this latter regard, it is probable that defense is a major beneficiary of implicit subsidies associated with non-payments for energy. 18 According to the Treasury Department of MoF, the defense accounting system is based on old accounting conventions developed before the turn of the century (under Tsarist Russia) and employs highly antiquated as well as a very non-transparent methodology. 19 This figure is from the attachment to Gov. Resolution (GR) 600, which goes on to indicate that 4.3 million people are deployed on federal legislative, executive and judicial governments, the procuracy and their territorial bodies, 1.9 8 time series data to determine how public employment changed. However, data on ‘state government management’ and general government employment available from GKS indicates that while federal government employment contracted between 1994 and 1997 (by nearly 5%), regional government employment grew by nearly 10%. 20 30. These numbers per se do not indicate whether there is excessive public employment or not, although the near doubling of general government employment between 1992 and 1996 reported by the State Statistics Agency is suggestive, particularly at the subnational level where growth has been most rapid. The picture, however, is mixed depending upon the part of government under scrutiny. In “traditional� public sector areas (e.g., health, education) there is fairly persuasive evidence of overstaffing; in new areas related to the functioning of a market economy (e.g., regulatory agencies, the judiciary), understaffing appears to be the problem. Recommendations for staff reduction, however, must be grounded in careful analysis of staffing adequacy within a strategic vision of the role of government and the efficacy of public service delivery. 31. While numbers appear to be excessive, compensation is modest. Wages at the federal level account for about 3 percent of GDP, or 25 percent of non-interest expenditures. The extensive provision of benefits in-kind for civil servants increases personnel driven costs. Free apartments, free medical clinic privileges, personal cars, subsidized meals, spa and resort services, and many other benefits distort the real costs of the civil service to the budget. 32. The civil service system and associated personnel costs thus merit careful consideration in an assessment of Russia’s public expenditure policies. Contingent Liabilities 33. Russia, like many other countries, does not budget for contingent liabilities. A comprehensive accounting of these potential claims on public resources is required for thorough evaluation of fiscal performance and risks to the fiscal accounts. There are several sources of potential obligations that may ultimately impose a cost on the budget but which do not appear in conventional presentations of fiscal accounts. Examples of contingent liabilities are loan or deposit guarantees, which only become a direct cost to the budget in the event of bank failures. Liabilities can also be either explicit or implicit: law or contract recognizes the former, the latter implies that the government has a ‘moral’ obligation to take over other’s liabilities. 34. Russia faces a number of such potential liabilities: • Explicit loan guarantees have been extended by the federal government on loans from foreign governments to Ministries and EBFs, and in some instances from commercial banks to finance budget operations. Information on the value of million work for budget supported organizations subordinate to the state and funded by the federal budget, and 309,000 work for other organs that are direct federal funds recipients. 20 GosKomStat, Russia in Figures, Moscow, 1998. 9 guarantees called is limited and appears modest (one source suggests 0.1% of GDP in 1998) 21 • Explicit guarantees on bank deposits have been extended to both the existing deposits of Sberbank, and to deposits transferred to Sberbank from seven large private banks severely hurt by the August 1998 financial crisis. Combining both existing deposits and those transferred, the total amount of guaranteed deposits is estimated at around 5.7% of GDP in 1998, including both ruble and dollar deposits. • Implicit (contingent) liabilities of the banking system constitute potentially a larger fiscal burden following the financial crisis in 1998, although small by international comparison with other banking crises due to the relatively small size of the financial sector in Russia. The public cost of restructuring and recapitalizing failed banks arises through direct and indirect assistance from the budget, CBR interventions, and indirect costs through lost revenues. Estimating the total costs to the public is very difficult but thought to be between 4% and 8% of GDP.22 • Implicit guarantees on subnational government liabilities. The value of outstanding subnational liabilities totaled at least R125 bn in 1998, or 5% of GDP. • Explicit budgetary arrears of the federal government have declined since the crisis, but remain very high for payables and receivables. The current stock of federal entities wage arrears has been reduced to R2.1 bn, arrears to enterprises is estimated at R20.6 bn while outstanding tax arrears owed the federal government stood at R218 bn as of Jul.1, 1999. • Extra-budgetary funds constitute a separate and potentially large source of explicit direct liabilities for the public budget. The largest is the Pension Fund (PF), with chronic financing problems and payment arrears as of Dec. 1998 totaling R[26] bn (0.9% of GDP), but completely disappearing by Oct. 1999. 23 35. These areas of explicit and implicit liabilities of government which are not formalized in the federal budget constitute a large source of risk to the federal budget and a risk to fiscal stabilization efforts. There is not adequate budgetary provisioning for these risks.24 Hence, when crises arise such as the 1998 banking crisis, it has a massive impact on public finances in ways that are still felt over one year later. The policy agenda for this fiscal problem should include explicit recognition and inventorying of full direct and indirect liabilities of the public sector, development of a legal framework for their management and oversight including firewalls where appropriate between the federal budget and other budgetary entities. 21 The 1999 draft budget law included a report on actual federal budget payments in 1998. Provisioning for called guarantees of R 1.5 bn was in the 1999 draft budget but subsequently removed by the Duma. 22 Standard & Poor’s “Sovereign Ratings Service�, October 1998. 23 Arrears were eliminated mainly through pension indexation far below inflation since August 1999, while the growth in the wage bill has boosted PF collection. 24 The draft 2000 federal budget includes Rb 0.65 bn to cover potential claims against earlier issued government guarantees, but this represents less than 5% of guarantees planned in 2000. 10