Report No. 20342-CHA China Managing Public Expenditures for Better Results Country Economic Memorandum April 25, 2000 Poverty Reduction and Economic Management Unit East Asia and Pacific Region FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS (As of March 31, 2000) Currency Renminbi Currency Unit = Yuan (Y) $1.00 = Y8.2 Y1.00 $0.12 FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES Metric System ABBREVIATIONS AND ACRONYMS ANAO - Australian National Audit Office DEET - Department of Employment, Education and Training DoF - Department of Finance EBA - Extra Budgetary Account EBF - Extra-Budgetary Fund ECD - Evaluation Capacity Development FM - Financial Management FMIS - Financial Management Information System GAO - General Accounting Office MOF - Ministry of Finance MOFTEC - Ministry of Foreign Trade and Economic Cooperation MTEF - Medium-Term Expenditure Framework MYEF - Multi-Year Expenditure Framework NAO - National Audit Office NCB - National Competitive Bidding NGO - Non-Governmental Organization NPC - National People's Congress NPO - Non-Profit Organization NPR - National Performance Review NTD - National Treasury Department OECD - Organization of Economic Cooperation and Development PBC - Peoples' Bank of China PEP - Portfolio Evaluation Plan SAT - State Administration of Taxation SOE - State-Owned Enterprises SPC - State Planning Commission SETC - State Economic and Trade Commission SRF - Self-Raised Fund TVE - Township and Village Enterprise VAT - Value Added Tax ZBB - Zero-Based Budgeting Vice President: Jemal-ud-din Kassum Country Director: Yukon Huang Sector Director: Homi Kharas Task Manager: Christine P. Wong FOR OFFICIAL USE ONLY TABLE OF CONTENTS Page No. Acknowledgments v Abstract vi CHAPTER 1: The Performance of China's Public Expenditure Management System........... Aggregate Fiscal Discipline.................................................................................................. Strategic Priorities................................................................................................................. 4 Efficiency of Service Delivery.............................................................................................. 5 Decentralization and Public Sector Performance.................................................................. 7 A View from the Top ............................................................................................................ 8 CHAPTER 2: OECD Experience in Public Expenditure Management Reforms..................... 10 The Background of OECD Reforms ..................................................................................... 10 Recent Reforms ..................................................................................................................... 12 Linking Policy, Planning and Budgeting............................................................................... 13 Redesigning Organizations.................................................................................................... 22 M arket Incentives...................................................................................................................23 The Changing Role of M OF.................................................................................................. 24 C o n lu sio n s ........................................................................................................................... 2 5 CHAPTER 3: Towards a M ore Com prehensive Budget............................................................. 26 Extra-Budgetary Funds ......................................................................................................... 26 An International Perspective on Extra-Budgetary Funds...................................................... 28 M inimizing Disadvantages of EBFs...................................................................................... 28 Reforming China's Extra-Budgetary Funds.......................................................................... 31 CHAPTER 4: Linking Policy, Planning and Budgeting.............................................................. 37 China's Incremental Budget Process..................................................................................... 37 Recent Budgetary Reforms ................................................................................................... 41 M aking the Budget an Effective Policy Tool........................................................................ 42 Linking Strategic Policy Choice and the Budget.................................................................. 44 M ove Away from Incrementalism ........................................................................................ 48 CHAPTER 5: Improving Incentives for Public Sector Performance......................................... 52 China's Decentralized Government...................................................................................... 52 Expenditure M onitoring and Control.................................................................................... 55 Recent Reform s..................................................................................................................... 57 The Changing Role ofM OF.................................................................................................. 63 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Annex 1: China's Expenditure Management System....................................................................... 65 Annex 2: Expenditure Management Issues Raised in Budget Speeches 1992-1997....................... 73 Annex 3: Expenditure Needs for Priority Areas .............................................................................. 79 Annex 4: China's Fiscal Prospects in the Medium Term................................................................. 84 Annex 5: Further Strengthening the Management of Extra-Budget Funds...................................... 95 Annex 6: Linking Planning, Policy and Budgeting in a Medium-Term Framework....................... 99 Annex 7: The Australian Experience with a Medium Term Expenditure Framework.................... 107 Annex 8: The Running Cost System in Australia............................................................................. 110 Annex 9: China's Intergovernmental Fiscal Relations..................................................................... 117 Annex 10: Expenditure Assignment: Theory and Some International Evidence............................. 131 Annex 11: A Formula-Based Equalization Transfer System for China: Model and Simulations... 140 Annex 12: Financial Management Information Systems.................................................................. 154 Annex 13: The United States: Federal Highway Trust Fund........................................................... 173 Annex 14: Evaluation Capacity Development in the Australian Government................................. 176 Annex 15: The United Kingdom's Next Step Agencies.................................................................... 196 B ibliography ......................................................................................................................... 197 R eferen ces............................................................................................................................. 200 BOXES IN TEXT Box 2.1 Earlier Budget Reforms in the OECD................................................................................. 12 Box 2.2 Excerpts from the 1997 Medium-Term Budget Policy Statement in South Africa............ 19 Box 2.3 The Netherlands Reconsideration Procedure....................................................................... 20 Box 2.4 Examples of Output Definitions and Costs in New Zealand............................................... 21 Box 2.5 Key Provisions of the Government Performance and Result Act of 1993.......................... 22 Box 2.6 Outcomes, Outputs, Programs, Inputs................................................................................. 23 Box 3.1: "W ould you like that w ith fees?"........................................................................................ 26 Box 3.2: Pros and Cons of Extra-Budgetary Funds.......................................................................... 31 Box 3.3: Japan's budget: General and special accounts................................................................... 35 Box 4.1: A n In-Y ear B udget R equest ............................................................................................... 39 Box 4.2: Mandatory Expenditure Increases in Laws and Decisions................................................. 41 Box 4.3: Zero-Based budgeting in Hubei Province........................................................................... 42 Box 4.4 The Changing Role of Government..................................................................................... 46 Box 4.5: Outline of China's Government Downsizing..................................................................... 48 Box 4.6: Intergovernmental Macroeconomic Coordination in Germany and Australia................... 49 Box 5.1: US and World Bank Procurement Rules............................................................................ 61 Box 5.2: The Four Pillars of Successful Civil Service Systems....................................................... 62 iii FIGURES IN TEXT Figure 1.1: China's A ggregate Fiscal D iscipline.............................................................................. 2 Figure 1.2: China's N onbinding B udget........................................................................................... 4 Figure 1.3: C hina: Strategic Priorities............................................................................................... 4 Figure 1.4: China: Efficiency of Services Delivery.......................................................................... 6 Figure 1.5: The Center's D eclining Share........................................................................................ 7 Figure 1.6: Expenditures Follow Revenues Follow Income............................................................. 8 Figure 2.1 Australia: Budget and Forward Estimates, Cumulative Real Growth............................ 18 Figure 3.1: D efinitions of G overnm ent............................................................................................. 36 Figure 4.1: Linking Planning and Budgeting.................................................................................... 43 Figure 5.1: China: Levels of G overnm en ........................................................................................ 52 Figure 5.2: The Bulk of Public Sector Personnel is in NPOs at the Local Level.............................. 53 Figure 5.3: Possible Structure of M O F ............................................................................................. 64 TABLES IN TEXT Table 1.1: General Government Expenditures.................................................................................. Table 1.2: Student-to-Teacher and Student-to-Total Staff Ratios..................................................... 6 Table 1.3: Central Government Share of Revenues and Expenditures in Selected Countries.......... 7 Table 2.1 The Growth of Government Expenditure, 1870-1990...................................................... 10 Table 2.2 The Size and Composition of Government Expenditures and Government Performance Indicators in Different Country Groups................................................................................ 1 Table 2.3 Public Expenditure Management in OECD Countries - Linking Policy, Planning and B udg eting ............................................................................................................................. 14 Table 2.4 Public Expenditure Management in OECD Countries - Incentives for Performance...... 15 Table 3.1: International Comparison of Extra-Budgetary Funds...................................................... 29 Table 4.2: Current and Reformed Budget Process............................................................................ 51 Table 5.1: China: Civil Service Wage Structure 1985 and 1993..................................................... 56 Table 5.2: A Perception of C orruption.............................................................................................. 57 iv p ACKNOWLEDGMENTS This report was written by a team including: Bert Hofman (team leader), Malcolm Holmes, Christine Wong, Zhou Xiaobing, Vargha Azad, Julia Li, Tamar Manuelyan Atinc, and Florence Parly (consultant). The report includes contributions by Jun Ma (IMF), JohnNorregaard (IMF), Heng-Fu Zou, Keith Mackay, and Maurice Gervais, and benefited from the comments of Vinaya Swaroop, Stefan Koeberle, Pieter Bottelier, Richard Newfarmer, William Allan (IMF), Russel Krelove (IMF) and the late Laslo Garamfalvi (IMF). Anne Green provided editorial assistance. Mmes. Leona Luo, Muriel Greaves, Meredith Dearborn, Grace Coward and Mr. Joseph Israel processed the report. The report was written under the guidance of Director Ding Xian Jue of the Comprehensive Department in MOF. The team would like to express its sincere thanks for the excellent cooperation the mission had during its stay in China. In particular, the team would like to thank Mr. DaiBai Hua and Ms. Wang Ling for their relentless efforts to make the mission a success. Mr. Dai and Ms. Wang greatly contributed to the report by sharing their knowledge, engaging in highly productive discussions, and by thoroughly preparing for the mission. Thanks also goes to Mr. Peng Long Yun, who helped prepare the mission and arranged for the very useful background papers. The team would also like to thank the Hubei provincial authorities and the Huangshi City and Daye County Finance Bureaus. A special thanks goes to Mr. Qiao of the Hubei Finance Bureau who was the perfect host during the visit. Finally, the team would like to thank the many Departments that gave the mission their time and insights: MOF Comprehensive Department, Budget Department, local budget departments, Agriculture Department; the Ministry of Agriculture, the Non-Profit Organizations (NPOs) and enterprises under the Ministry; the People's Bank of China (PBC) Treasury Department; the China National Audit Office (NAO), and the Standing Committee on Finance and Economics of the National People's Congress. The report was discussed with Government during a workshop in July 1998. The workshop was attended by Vice Minister Lou Jiwei, and officials from MOF, SAT, the State Council Reform Office, and others. During the production of the report, the Sector Managers were Klaus Rohland and Pieter Bottelier, and the Country Directors were Yukon Huang and Nicholas Hope. V 6 s , ABsTRA CT In China's emerging market economy, public expenditures will have a powerful impact on economic stability, social and economic development, and income distribution. China's continued success in these areas depends on how well it manages public expenditures. Achieving results with Government spending requires expenditure management that enforces fiscal discipline, directs spending to the Government's priorities, and ensures efficient delivery of public services. China's expenditure management does not yet meet these requirements. Over the reform period, China has kept budget deficits low, and has shifted spending to priority areas, and the country shows impressive economic and social outcome indicators, such as high growth, and high health and education standards. But China's public expenditure management faces profound challenges: * The retrenchment of the Plan puts more of the burden for macro and microeconomic policy on the budget; * Extra-budgetary funds and quasi-fiscal operations of the banking system undermine fiscal discipline, which contribute to China's repeated bouts of inflation. The overall fiscal deficit has been much larger than the officially recorded 1-2 percent of GDP, and could add up to as much as 10 percent of GDP; * Shifting spending to the Government's priorities is slow, and is in part undone during budget implementation. Many extra-budgetary funds may not support the Government's priorities at all. State Council priorities such as health, education, infrastructure and the environment are 4 to 5 percent of GDP underfunded, while SOE and bank restructuring put additional demands on fiscal resources. Over time, this could threaten sustainable and equitable growth; * While overall social indicators are high, regional disparities remain large. And government services seem overstaffed compared to other countries, which could escalate costs if wages continue to rise. To address these challenges, China needs to improve its public expenditure management. OECD reforms over the last 15 years have highlighted that changing the institutional framework-the rules of the game-for public expenditures is crucial for budgetary outcomes. Reforms focused on: rethinking the role of the State; shifting to multi-year budgeting; improving links between spending and policy priorities; and emphasizing performance in the public sector. The reforms paid off in lower deficits more focused spending, and better service delivery. In designing its own reform program, China should realize that these reforms take substantial time, resources, and a strong high-level political commitment. China's first priority is restoring fiscal discipline. Fiscal discipline requires including a significant part of extra-budgetary funds in the budget. However, Government should carefully balance the need for more supervision, and the incentives of budget units to deliver services and mobilize resources. Budgetary and extra-budgetary funds alike need better accounting, financial reporting, and treasury management. In addition, monitoring fiscal risk, and managing SOE guarantees and tax expenditures should reinforce fiscal discipline. But the most urgent task is to delineate a clear budget constraint for every line ministry and unit, and breaking down the sectoral budgets into organizational budgets is China's highest priority. vi To forge a stronger link between the State Council's policy priorities and the budget, China needs to revamp its budget process. Building on ongoing experiments, China could: (i) develop a medium-term expenditure framework; (ii) take decisions on policy together with decisions on spending; (iii) require detailed budget proposals that establish the link between proposed spending and China's priorities; and (iv) develop capacity for evaluating expenditures on their impact on these priorities. More openness in the budget process, including a larger involvement of the various People's Congresses, would enhance the incentives for policy makers to deliver results. But perhaps the most crucial step for China to take is to integrate capital and recurrent spending in one budget. The State Development and Planning Commission should then focus on articulating the Governments strategic priorities, but leave detailed planning for achieving these priorities to line ministries. In the short term, China could already make much progress by requiring agencies to submit savings proposals along with spending plans; avoiding in- year approvals of additional budget requests; and phasing out spending increases prescribed in laws. China's decentralized administration can be a major asset for cost-effective service delivery, if accountability for performance is improved. Beyond better financial management and audit, this requires better specification, reporting and monitoring of results of expenditures. Once accountability is established, more discretion for managers in allocating operating expenditures would improve results. China can further reduce the costs of public services by making wider use of the non-state sector through divesting from SOEs, relinquishing government direct control over many NPOs, and improving competitive bidding practices. Continued reforms in intergovernmental fiscal relations are essential for better public sector performance. The large provincial share in expenditures demands a clear assignment of responsibilities and closer coordination with the provinces to achieve macroeconomic stability and better outcomes on State priorities. But regional disparities in service delivery will only diminish, if the experimental equalizing grants scheme is substantially expanded, and earmarked grants are better targeted to the poorer provinces and counties. Over time, local tax autonomy matched with increased popular participation is the single best guarantee for well-managed local expenditures. vii CHAPTER 1 THE PERFORMANCE OF CHINA'S PUBLIC EXPENDITURE MANAGEMENT SYSTEM Over the reform period, China has managed to keep budget deficits low, shift budgetary expenditures toward priority sectors, and achieve remarkable gains in social indicators, despite a drop in the revenue to GDP ratio. However, extra-budgetary funds (EBFs), quasi-fiscal activities, and in-year budgetary adjustments undermine fiscal discipline; priorities remain underfunded, priority spending gets diverted during the budget year; and high regional disparities in social indicators persist. China's fiscal system has come under pressure development of the secondary market in the early over the reform period since 1978. The reforms 1990s. People's Bank of china (PBC) financing initiated a major shift of the tax base away from of the budget deficit remained below 2 percent of the sources traditionally tapped by the tax GDP, except for 1979, and stopped altogether system, and government revenues declined from after 1994. Thus, China appears to have over 34 percent of GNP in 1978 to less than 12 maintained an expenditure level consistent with percent in 1996. This chapter reviews how macroeconomic stability. China's budget system coped with the decline, in terms of maintaining fiscal discipline, Real expenditures increased. Fiscal adjustment meeting the Government's priorities, and appears less remarkable when real expenditures delivering services efficiently. are considered. While expenditures declined as a share of GDP, they rose considerably in real terms.1 Most governments would be envious of the 4 percent increase in real resources that Budget deficits remained modest. China's key Chi eperence overe reor e tf fiscal challenge over the reform period was to China hadike overnen expend.I remamn withmn the shrmnking resource envelope. Chn .a etGvrnetepniue reman wthi theshrnkig rsoure evelpe, constant in real terms for only one year during China's budget system therefore seems to have managed the declining resource envelope Table 1.1: General Government Expenditures remarkably well. Government budgetary Government Central expenditures and net lending as a share of GDP expenditures government as percent of expenditures as declined in line with shrinking government GNP percent of total revenues. Expenditures were less than 13 expenditures percent of GDP in 1996, well below that of All countries 39.1 72.3 comparator countries (Table 1.1 and Figure Industrialized 1.1). This remarkable reduction prevented the countries 47.6 65.9 Developing countries 31.7 77.8 budget deficits that caused much of the China (budgetary, 13.1 40.2 macroeconomic instability in other transition 1996) countries. Budget deficits have remained at a Source: MOF, Levin (1991), World Bankstaff estimates. Levin's data modest 1.5-2.5 percent of GDP throughout the are from a sample of 18 industrialized countries and 22 9 developing countries for which data on general government reform period, with the exception of the 5.3 are available in the International Finance Statistics. Data are percent peak in 1979. averages over three years ending 1987 or 1988. Bonds, reintroduced in 1981, financed a growing share of the budget deficit. Voluntary The real expenditure numbers are calculated with the GDP deflator. This may be an underestimation of the true inflation placement of bonds became feasible with the cost of Government services. 2 the 1990-95 period, it would have eliminated Y236 billion or 4 percent of GDP in 1995.2 With the budget deficit, a goal repeatedly expressed these funds included, overall government by the Finance Minister (see Annex 2). expenditures still declined from 37 percent of GDP in 1978 to about 21 in 1995, but in real EBFs rose sharply. Fiscal adjustment seems even less impressive, if EBFs are taken into terms, the annual increase averaged 6.5 percent. account. These funds consist of surtaxes, levies Moreover, 21 percent of GDP in government and charges accruing to government and expenditures is much more in line with comparator governments' levels of expenditures. Quasi-fiscal financing is large. The decline in Figure 1.1: China's Aggregate Fiscal Discipline China managed to keep budget deficits low... ... but real expenditures rose rapidly 400 Yuan 500 450 Real expenditure and net lending 200 350 oo 1978 1980 1982 1984 196 1988 1990 1992 1994 1996 19768 u 1.2 1 se4 1M S m 19 w 19 l 1l Note: Deflated with the GDP deflator. Wage deflator and RPI yield similar results. ....as did extrabudgetary and offbudget funds and the public sector deficit remains high 12 Social Insurance and 14 welfare 1 1 1 9 1 1 900Ext"budgetaryJ 1999 198. I984 19- 19., 197I8899 1s.91 192,0901 1999 9 1 .,9 99M 199 199 192 190 99 995 I99 * Extrabudgetary funds of government and administrative units Source: MOF, Budget Speeches, IMF, Statistical Yearbook. from 2.6 percent of GDP in 1978 to over 4 budgetary resources in terms of GDP forced percent of GDP in 1995, according to official China!s government to resort to the banks for statistics. A 1996 audit revealed that these financing policy expenditures such as funds could have amounted to over 6 percent of investment outlays and social expenditures of GDP, and later audits put the figure at 8-10 state-owned enterprises (SOEs). Much of this percent of GDP (Chapter 3). These funds are policy lending was directly or indirectly controlled by government, and have made up financed by the PBC: the PBC loaned to the for part of the decline in budgetary resources. State Banks, which in turn lent to SOEs. Social security funds, usually included in fiscal Conceptually, such loans are equivalent to resources in other countries, added another 2 State Statistical Yearbook 1997, p.747. Some double counting is possible as the budget provides limited contributions to these funds. 3 government net lending, and are usually replacing the bad debt with Government bonds. dubbed quasi-fiscal activities. Thus, the If this occurs, government debt would increase government's financing requirement was by 20-percentage point of GDP, for a total of significantly larger than the budget deficit about 35 percent of GDP. Although this is still suggests. The Consolidated Government sustainable in light of China's high growth rate, Deficit, which adds the budget deficit and a sudden increase in interest payments would central bank policy lending, was estimated to require painful adjustments in other budgetary average between 4.9 and 5.7 percent of GDP expenditures, or sudden increases in tax rates.' over 1985-1995.1 The Public Sector Deficit Moreover, if the banks turn bad, growth rates that consolidates government and SOE are likely to falter at the same time. Necessary borrowing is estimated to be 10.4 percent in tighter fiscal policy would add to lowering 1996 and well over 10 percent on average growth, which in turn could turn the between 1986 and 1995 (Figure 1.1). These Government's debt unsustainable., deficits, and their partial financing by the Central Bank, have contributed to China's The pension system is another potentially huge repeated bouts of inflation over the reform fiscal liability. The current system for SOE period. employees is heavily underfunded considering present benefit levels, premium payments, and Contingent liabilities could threaten budget demographics. Some experts estimate that the discipline. The bank loans that financed the "implicit pension debt" is over 50 percent of public sector deficit could become a GDP.7 Unless the pension policy changes, social government liability in the future. If a large security funds will run out of money, and number of SOEs are unable to pay back these government will face the awkward choice loans, widespread bank failure could result, between accommodating the budget demands, or and may force Government intervention to increasing premiums to a level that would protect depositors. Therefore, the non- severely damage labor market operations. performing SOE debt is a contingent liability Nevertheless, the pension debt is of a different for government. Official estimates put non- nature than the bad debts in the banks, because it performing loans at about 20 percent of the becomes due only over time. banks' portfolio, or about 20 percent of GDP. Because China's loan classification does not follow international standards, reality is probably grimmer.4 In a worse case scenario-, which is, still unlikely-government would have to bail out the banks, for example, by Even if a government bonds issue is not necessary, a more See Bert Hofman, Fiscal Decline and Quasi-Fiscal Response: gradual solution for the banks' bad debt is most likely to have China's Fiscal Policy and Fiscal System 1978-1994 in: fiscal consequences as well. For instance, banks could be Different Approaches to Market Reforms: A Comparison recapitalized with increased provisioning for bad debts. Between China and the CEECs, OECD/CEPII/CEPR 1997. However, this would reduce bank profits and therefore tax revenues. Altematively, lending rates could be set such that Loans in China are not yet classified according to risk, banks generate more profits. However, this would reduce borrower cash flow adequacy to service a loan, or to profitability of SOEs, the banks' main clients, and therefore probability of repayment. Rather, the existing classification reduce revenues to the budget as well. system is based on the number of days a loan is past due. All banks, under MOF regulations, are required to classify 6 Broad conditions for sustainable deficits require the growth rate outstanding principal or interest as (a) "overdue" if payment to exceed interest rate. These conditions may no longer hold, if has been overdue for more than six months but less than three growth suddenly drops. However, on the other hand, some taxes, years, (b) as a "collection loan" if overdue for more than three such as excises on petrol, could be quickly, and easily raised, and years, and (c) as a "bad loan" if it satisfies published MOF could eliminate the deficit. criteria. The amounts "classified" represent only the portion actually past due and not the underlying principal of the See China: Pension System Reform, World Bank report No. specific loan or the total exposure to a troubled borrower. 1512 1-CHA. 4 A budget that does not bind. China's budget market prices., However, shifting spending can does not constrain budgetary expenditures be slow: agriculture has been on the during the budget year. Since 1993, in every Government's list of priorities since the early year but one, expenditures were higher than budgeted. In the mid-1980s over-budget Figure 1.3: China: Strategic Priorities expenditures could add up to almost 20 percent Budgetary expenditures shifted to public goods of total (Figure 1.2), but even as recent as (expenditure shares 1978-1996, percent of total) 1993, excess spending was 12.5 percent of Percent of Total total. If Government had not spent beyond the 100-Capital Ependitures 90% 1995 and 96 budget, it would no longer have a 80% Subsidies budget deficit. 70% Agcuure culture. Education. 60% Science, Health 50% Strategic Priorities 40Defen Adrministration Shifting Priorities. Over the reform period, ote 20% China has gradually shifted budgetary 10% expenditures from economic construction to 0% 1 1978 1985 1995 1996 public goods provision (Figure 1.3). Within the Year shrinking budgetary envelope, the expenditure But strategic priorities get shortchanged during composition changed considerably, reflecting the year economy-wide reforms. For example, the (in-year overspending, percent of budget) separation of enterprises from the budget caused a strong decline in capital construction n. expenditures. This decentralization of w i. responsibility was crucial for achieving China's spectacular growth rates. Price Figure 1.2: China's Nonbinding Budget Edun (In-year overspending, Yuan and Percent of budget) A-0- rer~1Percent Source: Various budget speeches, MOF eighties, but got a declining share of the budget until 1995. Similarly, despite the emphasis on infrastructure since the mid-eighties, spending only caught up after 1992. Despite the shift in expenditures, priority areas Source: Budget Speeches such as health, education, poverty and the reforms pushed up subsidies for daily living environment still remain underfunded. Provided until the mid-1980, and after 1985, rising SOE that services are delivered cost-effectively, these subsidies feathered off the impact of producer areas may require an additional 4-5 percent of price reforms. In recent years the budgetary GDP in expenditures (Annex 3). Although such share of both types of subsidies declined as estimates can only be preliminary, they are consumer and output prices were aligned with 0 Although the 1992-1996 inflation was combated in part by reintroducing consumer subsidies for grain. 5 probably in the right order of magnitude. Uneven distribution. China's overall high Underfunding in these areas could, over time standards in service delivery mask dramatic undermine sustainability and equity of China's regional disparities. Beijing has the same infant growth. They also signal the problem that mortality rate as France, but Qinghai and China cannot both finance its priority Yunnan are barely equal to Ghana or Bolivia. expenditures and eliminate its deficit, without Similar disparities can be found in other health reallocating expenditures within the budget, or and in education indicators. Of course, public taking revenue measures (Annex 4). services are not the only factors in these outcome disparities: regional per capita income Whether EBFs have been reallocated towards and large population of ethnic minorities are Government priorities cannot be established. important factors as well." The accounting and classification system of these funds differs from budgetary funds, and No sight on cost-effectiveness. Assessing reporting on these funds is often late and whether China's government services are irregular. delivered at minimum costs is not possible with currently published data." Not only is it In-year adjustment. Priorities get diluted impossible to aggregate budgetary and EBFs, during the budget year. All budgetary and significant off-budget funds used for expenditures are substantially higher than financing government services, but also publicly budgeted, but there is significant available expenditure classification reveals too differentiation among the sectoral increases little detail to break down expenditure by major (Figure 1.3). Priority areas such as education program or by cost center and frequently even and agriculture get shortchanged during the the Ministry of Finance (MOF) does not have budget year, whereas enterprise support access to relevant cost information. through working capital and technological transformation expenditures benefited from in- Too many government workers? The scarce year allocation of additional revenues. available data suggests that government services are delivered with high levels of labor inputs. Efficiency of Service Delivery Overall, some 5.3 percent of the population Remarkable outcome indicators. Outcome work for China's government, compared to 5.7 indicators for China's public expenditures look percent in the Organization of Economic good in comparison with other countries at Cooperation and Development (OECD) good. countries, 3.6 percent in developing countries, similar levels of GDP (Figure 1 .4). Indicators such as literacy, infant mortality, life and 2.6 percent in Asian countries." and malnutrition are usually Overstaffing is quite striking in university exphety teducation: China stands out with only 2.7 mouc t er su tng t el of comparat students per staff, compared to 21 in Korea, but countries, suggesting relatively efficient ony3.ecn fCiasuiest tf service delivery in education, health services, and water and sanitation. Output indicators of these and other government services also indicators, see: Pradhan, Sanjay: Evaluating Public Spending, thes andothr goernent ervces lso World Bank Discussion paper No.323, May 1996. compare well to other developing countries: c See China: Regional Disparities, World Bank Report -14496- primary and secondary enrollment rates, CHA (Green Cover). access to sanitation and safe drinking water, "The mission that prepared this report had only limited access to and access to paved roads are usually at levels non-published sources, not least because much of the relevant similar to middle income countries., information was not reported to MOF, or because reporting formats did not allow for analysis of the data. 12 See Sciavo Campo et al. An International Statistical See World Development Indicators, 1997; and China: Public Comparison of Government Employment and Wages, World Investment and Finance, World Bank report No.14540-CHA, Bank Policy Research Working Paper No. 1806, 1997. October 1995. For a discussion of outcome, output, and input 6 Figure 1.4: China Efficiency of Service Delivery China's social indicators are better than most conparators Illiteracy ratio 7o75 SFale E Mal. (n1ani VictNa= 40 4 agol * InFinip ne 4(1 1 ww, Wlmi 55 *NTp1 -~IO .,i'. 101 0 500 1000 1500 2000 2500 Bat the spoils are spread unevenly among provinces Ilteracyl Perut 1sai) Infant mortaltyll.000 70 8Dbirths 00 60 Qingha 40 50 0uzho. 40 utvnii94 20 10 L 0 10 ** Shangai 0 1000 2000 3000 400D 50 0000 7000 8000 90 10000 0 Per Capita Inaome 0 2000 4000 6000 800D 10000 1993 Pe Capita incom, 1993 And resaults require high labor inputs that can become costly Tehes pra prim.ay choI PhysicA., I1bpit.l student Brds/10,0409pople thiadMortco 1 Hospital Bed I (91 I hi4-b 11 101 nu 50 DD 1500 2000 250D 3(00) 350 0 D Income per Capita, 1995 ...ae 0. .ra 014.o.. U,O.. ISk,on4... . runnu4 v.1.. Sorce: MOF, State Statistical Yearbook 1995, 1996; World Development Indicators 1997 Table 1.2: Student-to-Teacher and Student-to- Total Staff Ratios teaches (Table 1.2). The high staffing levels Full-time Student-to- Student- could in part be explained by relatively low teachers as full-time to-total % of total teacher staff wages that could make it economical to have staff ratio ratio low student/teachers and population/physician ratios. The total public sector wage bill in 1996 China (1994) amounted to Y220 billion, or little over 3 * Regular Higher 38 7:1 2.7:1 percent of GDP, which is small by Education international comparison. However, wages are North American & -- 15-20:1 7:1 European Universities likely to increase more rapidly than GDP in the Taiwan (China) (1992) 75 21:1 16:1 Republic of Korea (1994) 66 33:1 21:1 Source: UNESCO, Statistical Yearbook 1995, pp. 1-30-1-51 7 future as nonmonetary benefits are Local governments receive the bulk of EBFs.1 monetized.,` Some estimate these benefits to Over 85 percent is locally administered, most of equal 50 percent of pay. Thus, the current it by the administrative units and institutions input structure could become too expensive. under local government. Social security funds are also locally administered, and pooled usually Decentralization and Public Sector at municipal levels. Equally, revenues from Performance land leases and SOE divestiture-funds not recorded in any published account-accrue The performance of China's public mostly to local government. expenditure management system can only be fully understood with a thorough understanding of its intergovernmental fiscal Table 1.3: Central Government Share of relations (Annex 10 provides a detailed Revenues and Expenditures in Selected account). Central government's expenditure Countries share declined for most of the reform period, (percent of total) as did its revenue share (Figure 1.5). In 1996, Year Revenues Expen- central government administered only 27 ditures percent of all expenditures, and even after the Federal countries radical 1994 reforms, received only half of all Australia 1995 76.6 59.0 Canada 1993 53.5 41.7 revenues. Central government's share is now Germany 1995 73.0 59.2 much lower than in other developed and Spain 1993 86.6 70.4 developing countries, particularly on the United States 1994 65.7 53.5 India 1993 61.8 54.7 Figure 1.5: The Center's Declining Share Argentina 1992 57.2 55.1 Brazil 1993 71.4 65.7 (Central expenditure and revenue share 1978-96) Mexico 1993 84.6 78.3 PercentofTotal Russian Federation 1995 60.0 62.4 00 Unitary countries 50 D, Expenditures Belgium 1994 94.8 88.5 France 1995 89.8 82.3 Netherlands 1995 96.3 76.4 wo Norway 1994 78.6 68.4 United Kingdom 1995 96.4 77.3 Kenya 1994 97.8 96.1 100 Revenues Poland 1995 92.1 83.8 South Africa 1994 91.4 66.3 00 Thailand 1995 94.9 92.6 1978 1980 1982 1984 1986 1988 1990 1992 1964 1996 1984 Rdfom in 1994 F'8Oal Enlerpdse Tax Refor Source: Ter Minasisan, T.:"Decentralizing Government" in Finance & Development, September 1997. Source: MOF, Liu Zhongli 1997. Local governments' large share in fiscal revenue side (Table 1.3), although part of the resources means that their expenditure difference can be explained by China's vast management is decisive for China's fiscal size." outcomes. Local governments are not allowed to borrow, but their ownership of most of the * Wages in China have lagged behind GDP growth over the SOEs puts much of the public sector deficit on reform period, and the official wage ratio of 20 percent of GDP is extremely low. However, much of workers' compensation is in the form of non-monetary benefits, such into account per capita income, urbanization, and size in terms as low-cost housing, education, health care, and underfunded of population or surface, China did not seem out of line with pension rights. If these benefits were to be monetized, a rapid others on the revenue side, although the expenditure side was rise in wages would be expected, significantly lower than expected. * See China: Budgetary Policy and Intergovernmental Fiscal 's The report follows the Chinese convention that designates all Relations World Bank report I1094-CHA, July 1993. Over a subnational governments as local. sample of 39 developing and developed countries, and taking 8 their accounts." In principle, the credit plan expenditures closely follow the highly dispersed prescribes the overall-borrowing limit of revenue base (Figure 1.6). The reason for this is SOEs, but in the past, local government that-except for Tibet-transfers from central to pressures have repeatedly led to adjustment of provincial government do not benefit the poorer provinces. Most of the transfers are "tax return" Figure 1.6: Expenditures Follow Revenues transfers, a consequence of the 1994 reforms in Follow Income intergovernmental fiscal relations. These (Provincial per capita revenues transfers redistribute the increase of central tax and expenditures, 1996) revenues since the 1994 reforms to the provinces in which the revenues accrued. Thus, Perc ... rafter Tibet, the richest province, Shanghai, .50receives the highest per capita transfer. 2000 However, other, more discretionary transfers, including earmarked grants and equalization S Eetgrants, are not targeted to poorer provinces. Revnue Trend 1000 A View from the Top S'_ Expendit_res China's policymakers are well aware of many 0 5000 1000 150o 200M0 25= public expenditure issues. Every year, the Per capita Ineme, Yuan 1900 budget speech enumerates a variety of issues Unequalizing transfers (Annex 2), which broadly cover each of the three aspects of public expenditure performance. (Per capita tax rebate and other transfers, 1996) the b sp eece eur e outathe The budget speeches regularly single out the Per capita transfer rapid growth of expenditures, in particular f000 administrative expenditures, and emphasize the 100 need for increased macro control. The speeches 1400 01O0o oransfers express the need to readjust the pattern of 1200Tax rebate 1o00 hansfers expenditures towards priority areas such as 8oo agriculture, education and science, poverty 00 reduction, and adjustment of the economic oa2 structure. Unfortunately, the failure to so, and 0i Jthe "diversion of funds" away from priorities is 2025 3854 5325 11629 also a recurrent theme. Finally, each year the Pr a..p.- comY. Yn Finance Minister addresses the "extravagance Source: MOF. and waste" and uncontrolled social consumption the plan, and even unauthorized overshooting such as overseas trips and banquets, indicating of the plan.,, that services could probably be delivered more efficiently. The disparities in expenditure outcomes are closely related to the disparity in fiscal Over the last two years, expenditure resources. Over the reform period, provinces management reforms have become an explicit relied increasingly on their own resources, and theme in the budget speeches, particularly the need to strengthen management of EBFs, and to "restore the financial order." These words have * See China: Budgetary Policy and Intergovernmental Fiscal been matched by actions: the 1994 budget law Relations World Bank report 11094-CHA has been a major step forward for budgetary * See China: Decentralization and Macroeconomic Stability, management by clearly defining the roles and World Bank report, October 1994. 9 responsibilities in the budget process. State Council Document No.29 of July 1996 has made a first step towards integrating EBFs in budgetary management. And budget speeches pay increasing attention to issues of social security and SOE reforms. These are promising beginnings. However, this report argues that for fiscal policy to become a powerful tool for China's government-as it is for OECD governments-a more fundamental overhaul of the budgetary system is necessary. The OECD experience in budgetary reforms will be reviewed in the next chapter. Chapters three, four and five scrutinize China's budgetary institutions and recommend changes in China's budgetary framework that will deliver more fiscal discipline, achieve the government's strategic priorities, and increase the efficiency of service delivery. 10 CHAPTER 2 OECD EXPERIENCE IN PUBLIC EXPENDITURE MANAGEMENT REFORMS Recent reforms in OECD public expenditure management reacted to rising fiscal imbalances and lagging public sector performance. The reforms have sought to improve macroeconomic and microeconomic aspects of the budget by better linking policy, planning and budgeting within a multi-year budget framework. Hard and predictable budget constraints and greater managerial autonomy matched with increased accountability have been decisive for better public sector performance. The Background of OECD Reforms the belief that government could solve many of Growing Governments. OECD reforms began the macroeconomic ills of the prewar decade, against a background of concern over public and needed to mend many of the market's expenditure growth, rising deficits and debt perceived failures. By the start of the 1980s, burdens, and a general disappointment with OECD governments spent over 40 percent of government performance. Governments had GDP on average, more than double the share grown rapidly after World War II, backed by of before the war (Table 2.1). Table 2.1: The Growth of Government Expenditure, 1870-1990 (percent of GDP) Later 1 9th Pre-World Post-World Pre-World Post-World Century War I War I War II War II (about 1870)/a (about 1913)/a (about 1920)La (about 1937)/a (1960) (1980) (1990) Austria .. 14.7 15.2 35.7 48.1 48.6 Belgium .. .. .. 21.8 30.3 58.6 54.8 Canada .. ... 13.3 18.6 28.6 38.8 46.0 France 12.6 170 27.6 29.0 34.6 461 49.8 Germany 10.0 14.8 25.0 42.4 32.4 47.9 45.1 Italy 11 9 11.1 22.5 24.5 30.1 41.9 53.2 Japan 8.8 8.3 14.8 25.4 17.5 32.0 31.7 Netherlands 9.1 9.0 13.5 19.0 33.7 55.2 54.0 Norway 3.7 8.3 13.7 ... 29.9 37.5 53.8 Spain .. 8.3 9.3 18.4 18.8 32.2 42.0 Sweden 5.7 6.3 8.1 10.4 31.0 60.1 59.1 Switzerland . 2.7 4.6 6.1 17.2 32.8 33.5 United Kingdom 9.4 12.7 26.2 30.0 32.2 43.0 39.9 United States 39 1.8 7.0 8.6 27.0 31.8 33.3 Australia . ... .. . 21.2 31.6 34.7 Ireland .. ... . 28.0 48.9 41.2 New Zealand /d .. .. ... 26.9 38.1 41.3 Total Average 8.3 9.1 15.4 20.7 27.9 42.6 44.8 La Or nearest available year after 1870, before 1913, after 1920 and before 1937. /b 1992. /c Average, computed without Germany, Japan and Spain (all undergoing war or war preparations at this time). /d GFS data, data available for 1960 is 1970, central government data only. Source: Schuhknecht and Tanzi (1996). 11 Disappointing results. Much of the growth in countries may also have contributed to lower expenditures resulted from social security, crime rates in countries with larger interest on rising debt levels and from govern- governments. In addition, part of the increase ment consumption, driven by rapidly in government expenditures can be explained expanding payrolls. Some therefore concluded by the lagging productivity inherent in that this expenditure growth contributed little government services, which automatically Table 2.2: The Size and Composition of Government Expenditures and Government Performance Indicators in Different Country Groups Industrialized Countries Newly "Big" "Medium-sized" "Small" industrialized Governments /a Governments /b Governments /c countries /d 1960 1990 1960 1990 1960 1990 1990 Total Expenditures (percent of GDP) 31.0 55.1 29.3 44.9 23.0 34.6 18.2 of which Health 2.6 6.6 3.0 5.9 2.3 5.2 3.3 Education 4.5 6.4 2.9 5.6 3.4 5.0 3A4 Social Security 13.5 19.5 9.6 13.9 6.2 7.9 1.0 Research and development ... 2.0 ... 1.6 ... 2.0.. Economic and Regulatory Efficiency Indicators Real GDP growth (in percent) /e 3.2 2.6 4.0 3.3 4.6 3.3 6.2 Gross fixed capital formation (in percent of GDP) 23.4 20.5 21.1 21.3 19.6 20.7 31.2 Inflation (in percent) 1.7 5.4 1.6 4.3 2.3 6.1 15.3 Unemployment rate (in percent) 2.9 6.1 4.6 9.2 2.7 4.2 2.9 Size of shadow economy (in percent of GDP)I/f 4.9 11.1 3.8 8.2 3.5 6.2.. Patents/I10,000 population (inventiveness coefficient) ... 2.0 .. 2.3 ... 8.6.. Social Indicators Rank in UN human developmentZ9 ... 11.0 ... 13.0 ... 6.0.. Income share of lowest 40 percent 15.6 24.1 16.4 21.6 17.4 20.8 17.0 Illiterate poptulation as percent of population 15+ 9.3 2.9 13.3 4.6 2.2 0.5/h 9.2 Secondary school enrollment (in percent) 55.0 93.0 51.0 99.0 61.0 89.0 81.0 Life expectancy 72.0 77.0 70.0 77.0 71.0 77.0 74.0 Infant mortality/'000 births 23.0 6.7 29.0 7.1 22.4 6.4 9.8 Prisoners/100,000 people ... 38.0 ... 68.0 ... 154.0/i.. Divorces (in percent of marriages contracted, 1987-9 1) ... 33.0 ... 33.0 ... 36.0.. Emnigration (in percent of total population)j 0.6 0.2 0.3 0.8 0.2 0.1 0.1 Lt Belgium.11 Italy, Netherlands, Norway, Sweden (public expenditure more than 50 percent of GDP in 1990). /b Austria, Canada, France, Gennany, Ireland, New Zealand, Spain (public expenditure between 40 and 50 percent of GDP in 1990). c Australia, Japan, Switzerland, United Kingdom, United States (public expenditure less than 40 percent of GDP in, 1990). /d Cle, Hong Kong, Korea, Sintgapore; 1990 or nearest available year. le Average of preceding five years, 1956-60 or 1986-90. /f Most recent data available is 1978, used in 1990 coluns. Lg 1992. IitUS only. Others below 5 percent, UNESCO statistics for 1991. TlExcluding United States, average is 64. /jData available for 1960 is 1970, data for 1990 may inclutde 1993 in some countriet. Newly industrialized country data, only Korea is available (1993). .Source: Schuhknecht and Tanzi (1996). to welfare (Schuknecht and Tanzi, 1996). The increases the government's share as income critics argued that increasing expenditure rises.' Nevertheless, the growing perception share did not "buy" more in terms of better by the end of the 1970s was that government outcomnes of government policy: most of the was the problem rather than the solution. economic and social indicators seem unrelated to government size. However, "Big IThis so-called Baumol effect argues that government would Governments" perform better in income need an increasing share of GDP if productivity increases in distribution and containing crime (Table 2.2). government services lags behind that of other sectors of the Higher spending on social security in these economy. 12 Box 2.1: Earlier Budget Reforms in the OECD Budgetary procedures had evolved from the end of the last century: * Line iten budgeting established previously lacked executive control over spending by ministries and agencies. In line item budgets, expenditures are listed according to objects of expenditures or line items, and focused on expenditure control, with MOF acting as controller. Line item budgets were not compatible with the expanding role of government. They gave no information on what the money was spent on or whether the money was used cost-efficiently or effectively. * Performance budgeting introduced measures of workload and cost of activities into the budget. This increased the focus on cost-efficiency, but still lacked effectiveness information. * Program budgeting introduced focus on trade-offs among competing goals. While performance budgeting focused on achieving a given goal at least cost, program budgeting treated the goals themselves as variables. Program budgeting requires the budget to be organized in groups of activities towards a certain goal, a program, and in addition required effectiveness and outcome measures. Program budgeting failed to become the main budget tool in part because the trade-off among competing goals could not be captured in the process, in part because the information required to make the trade-off was often not available because information systems did not support the budgeting technique. * Zero Based Budgeting (ZBB), introduced in the 1970s, focused on the process of budgeting rather than the contents. In a pure ZBB system, all programs are evaluated from scratch each year. This pure form has never been introduced, but many governments have used the principles of ZBB, by requiring ministries to propose budgets with only 90 or 80 percent of the existing allocation. The failure of performance, program, and zero-based budgeting to live up to their expectations was, in OECD countries, partially due to the high degree of centralization of the budget process that went along with them. As a consequence, the central ministries became overwhelmed with the paperwork involved in annual budgeting. In developing countries that introduced these or similar budget systems, the failure to address the generally disabling environment for performance in the public sector was key to the limited success. Ineffective budgetary procedures. Budgetary * Focus on distributing the fiscal increment procedures and practices in the 1970s across new spending proposals, and an contributed to expanding governments, and extensive and detailed debate on existing reinforced the perception that government was spending, but little discussion of new and incapable of delivering results. Despite earlier existing policies. reforms (Box 2.1), at the outset of the most recent reforms, OECD budgeting systems had * Few incentives for agencies to save all or most of the following characteristics: budgetary resources, because current year spending was the starting point for the * Primary focus on inputs, with performance annual budget haggle. judged largely by how closely spending matched budget appropriations. * Little clarity of purpose, task, and costs of policies, programs and services, and an * Short-term horizon for budget decision- intertwining in a single agency of policy making which failed to adequately account advice, regulation, service delivery and for longer-term costs and led to biased funding. choices of policy instruments, for example between capital and current spending and In sum, the budget processes in OECD between spending and regulating. countries created a disabling environment for performance in the public sector. * Bottom-up approach to budgeting, which created strong upward pressure on Recent Reforms expenditures. The arbitrary expenditure cuts in budget implementation that The OECD countries have adopted varying followed undermined program and agency degrees of reforms over recent years. Some, level performance. such as New Zealand, the UK and Australia, have adopted radical new approaches to expenditure management and public 13 administration. Others, such as Germany and Linking Policy, Planning and Budgeting Japan, have stayed much closer to the traditional, strongly bureaucratic model, which The Scope of the Budget once dominated budgetary management in OECD countries. These differences are not Off-budget operations crippled financial surprising: budgetary institutions and discipline in various OECD countries in the organizations are deeply embedded in history 1970s. Implicit liabilities in social security and culture, and are not easily transferred from systems and loan guarantees for (usually) one country to the other. Despite the SOEs often became explicit during economic differences, common strands in reforms exist. recessions that also suppressed government Most prominently among them is the focus on revenues. Off-budget funds also undermined achieving aggregate fiscal discipline, funding efficient service delivery, as agencies had strategic priorities, and delivering services recourse to alternate means of finance if they efficiently. Whereas earlier reforms had ran out of budgetary funds. addressed each of these individually, more recent reforms have taken a more integrated OECD budgets now include almost all approach and have recognized that these three expenses of government departments and their levels of performance are inseparable. agencies, whatever their source of funding. Exceptions arise only when a law treats a The OECD reforms aimed broadly to: particular agency as off-budget (Chapter 3). Social security funds may be separately * better link policy, planning and budgeting managed, but they are normally included in the by: (i) increasing the scope of the budget budget reports, and social security policy is to include expenditures for all government included in annual budget deliberations. Most activities; (ii) adopting global budgetary countries treat state enterprises only on a net targets; (iii) introducing a multi-year basis, reporting taxes and subsidies in the perspective to policy and budgeting; and budget. However, some-such as the UK and (iv) focusing on strategic policy decisions France-consider SOE borrowing during at the center of government, while budget preparations, and report on public delegating implementation to sector deficit and borrowing requirement. implementing agencies; and Many OECD countries have increased budget * improve performance by: (i) enforcing coverage to include tax expenditures and loan hard budget constraints of a guarantees. These are reported in the budget comprehensive budget; (ii) creating presentation or in mandatory regular reports to agencies with more focused tasks; (iii) parliament, such as the "subsidy report" in devolving managerial authority supported Germany, which is issued every two years. by better specification and measurement of New Zealand has the fullest treatment of performance; (iv) improving information "fiscal risks, "including contingent systems and audit; and (v) introducing liabilities-claims against the government market mechanisms in the public sector. which will arise if a specific event occurs-and The main features of some OECD public any event which if it occurred, would effect expenditure management systems are the revenues of the State.2 In some countries, summarized in Tables 2.3 and 2.4. including the United States, there is a regulatory budget so that the costs on the community of regulations are made public. 2 Underpinning the New Zealand approach has been a shift to a full accruals approach to budgeting. 14 Table 2.3: Public Expenditure Management in OECD Countries - Linking Policy, Planning, and Budgeting Australia France Germany Japan New Zealand Unites Kingdom United States Strategic Goals/ Commitment to zero budget Medium-tenn targets is the Financial Planning Council Reduction in bonds issue Move toward surplus inl Expenditure growth lower than Balanced budget by 2002. Global Budget deficit over the business "lois d'orientation' canl set expenditure limits to below 5 percent of adjusted financial balance; GDP growth. Targets cycle. Regular publications Maastricht criteria. Currently: Maastricht expenditures. reduce expenditures to GDP. of fiscal outlook. criteria. Targets in Fiscal Responsibility act. Multi-annual Forward estimates (budget Experimental 3-year rolling Five-year rolling plans No formal system of multi- Three year projections of Budgetary policy set within Office of Management and Budgeting year and three outer years) plans, annexed to the budget developed by MOF oni year estimates. Medium- "baseline" based on previous Medium-tenn Financial Strategy, Budget makes five-year central to budget process. documents. basis of Financial Planning term projections included year's numbers and with key focus oin borrowing projection, but no Council agreement, which in Budget Document. Government decisions; requirement. Presented in Budget operational significance. coordinates policy with issued to line ministries at Documents. New Control Total subnational Laender; the start of the budget drives multi-year expenditures. included in the budget process. Included in the documents, but no legal or budget bill. Outer year operational status. projections become baseline. Salient Features of Cabinet detennines overall Prime minister sets sectoral Ministries submit proposals Centrally imanaged by the Output budgeting. Budget Budget made within limit of Budget cycle divided in Budget Process expenditure limit, and organizational to MOF within limits set in Finance ministry, Budget based on baseline estimate control figures. Treasury invites administrative and Expenditure review expenditure limits based on budget circular. Office: Line ministries that includes policy "position papers" from congressional cycle. committee(ERC) considers MOF's internal projections. Submissions reviewed by submit proposals to MOF decisions. Ministers make departments, including spending Administration prepares on policy changes, based on Budget proposals should in MOF at various levels and ott basis of a cabinet decisions on output increases that need to be matched basis of OMB guidelines. policy proposals and savings principle not exceed limits, in the Cabinet, which approved circular, and purchases rather than with savings proposals. On basis Expenditures divided in options. ERC sets sector but Ministries may appeal to submits to Parliament. negotiate with MOF. spending. Spending of position papers, Cabinet decides mandatory and ceilings for individual Prime Minister. Two rounds of review in MOF seids proposal to ministries prepare guidelines for budget. Cabinet discretionary, the latter ministries. Ministries Parliament. spending ministries, appropriation estimates committee reviews ministerial subject to legal limits on the prepare sector budgets. followed by new based on Cabinet output proposals and proposes allocation deficit. President submits Program management negotiations. Cabinet decisions. of control figures. Cabinet proposal to Congress, focuses on programs as the decides. approves. which, after a complex basis for resource allocation. process, approves the budget. Budget Year, Budget July I-June 30: Preparation January 1-December 31. January 1-December 31. April 1-March 31. July I-June 31. April I-March 31, October 1-September 30. Calendar January-June Preparation January- Preparation starts 13 Preparation starts in April Process starts July of Budget process starts April of Budget process starts 17 December of previous year. months in advance. of previous budget year. previous year, previous year. months in advance. Expenditure Review ERC has reviews available No formal procedure. Court Ad hoc reviews of existing Efficiency and Annual budget cycle main "Efficiency scrutinies" to lower Evaluation of existing called for in previous budget des Comptes (audit office) policies. Biannual report effectiveness reviews in vehicle. Ad hoc reviews of costs; budget process forces to programs included in round. All programs performs analysis. on subsidies. budget process. Sunset specific areas. propose savings; Ad hoc reviews ministerial budget requests, evaluated over a 5-year legislation provides of major policy areas. OMB and CBO reviews of cycle. automatic review. specific areas. Powers of MOF or Responsible for economic, As in Australia, but with Responsible for fiscal MOF responsible for As in Australia, but with As in Australia; recently some Office of Management and Treasury in Budget fiscal and monetary policy. more detailed expenditure policy. Drafts and budgetary policy, tax more emphasis on monetary policy responsibilities Budget and Treasury have Preparation Manages the three-year control. proposes the budget; administration, banking scrutinizing budget proposals delegated to Bank of England. responsibilities similar to forward projections drafts evaluates and reviews supervision, Fiscal and administration. Australia, and proposes the budget; government programs and investment and loan evaluates and reviews expenditure proposals; in program. government programs and charge of financial expenditure proposals; in administration. charge of financial administration. Role of Parliament No power to propose new No power to propose Unlimited powers to come Both houses of the Diet No power to come up with No power to come up with new Unlimited powers to come policies; may amend the spending. Parliament may sup with new proposals or can make amendments and new proposal. Executives' spending proposals. Taxes and up with new proposals and executives' proposal without raise expenditures if others propose amendments to the additions to expenditures proposals may be amended spending may be reduced, but not to amend executive increasing the tax burden, are cut executives' proposal. proposed. without raising the tax increased. proposals. may refuse approval of burden. budget. 15 Table 2.4: Public Expenditure Management in OECD Countries - Incentives for Performance Australia France Germany Japan New Zealand Unites Kingdom United States Performance Expected Outcomes and Line item budgets. Line item budgets. Line item budgets. Output budgets, strategic Output Budgets. Legislative Specification associated outputs Ministry and plan, business plan. requirement for included in budget Centers of strategic plan, documents. Responsibility agency Contracts. performance report. Budgeting for Ministries/agencies not available Unified personnel Total Staff Number Allocations for outputs to Personnel costs Total federal Operational receive single running system for central and Law determines be produced. Chief controlled by running employment Expenditures cost allocation, which is local government. maximum overall executives of ministries, cost system. Pay controlled; limits formula driven, and Number of posts staff. Cabinet order agencies have authority to increases in principle on number of requires efficiency approved in budget. and budget hire and fire. Limits set on financed from senior staff. dividend. No personnel actual number determines staff chief executive salaries, efficiency gains. limits on ministries and determined by numbers for agencies. Possibility to appropriation, but ministries. MOF shift resources between reserves can finance involved in salary personnel and running staff up to approved budget preparation. cost, and between number. program. Expenditure Control MOF controls MOF issues Approval by MOF is Quarterly Departmental accounts in a Warrant issued by the Warrant issued by aggregate cash limits; warrants and required before commitment and commercial bank. Minister of Finance. the treasury. expenditure controls checks in principle disbursement; disbursement plans Treasury clears the The Treasury, and Office of responsibility of line all expenditures, commitments can be approved by MOF. accounts every night, chief financial officers Management and ministry or agency; through Inspecteur blocked if a certain Government consolidates balances at the in agencies, ministries Budget, Treasury, MOF provides de Finances. percentage of consolidated account Crown Account in the responsible for and heads of accounting services; Recently has appropriation is spent; in Bank of Japan; central bank. The law expenditure control. department Single treasury account. devolved some MOF and heads of subaccounts for line gives MOF or Treasury a Single treasury responsible for Virement is possible authority; single agencies responsible for ministries. Virement possibility to require any account; paymaster expenditure between subheads and treasury account at expenditure control. within one item with information, which is used accounts for ministries, control. Single within running costs, central bank, Single treasury account approval of MOF. as a base for a warrant Virement between treasury account; however, this is not encompassing encompassing regional Among items with system. Transfer allowed subheads is possible Tax and loan regulated by the law. regional accounts. accounts. Virement Parliamentary up to 5 percent from output but needs approval by accounts in Carryover up to 6 Virement is possible within the same approval. to output, if no other the treasury. The commercial percent of running possible within a chapter, and between Carryover automatic, transfer to the same Treasury has no power banks, on call for costs; capital limit of 10 percent salaries and wages. if expenditures appropriation has been to authorize virement Treasury. The expenditure may be of the appropriated Must be approved by approved as made during the year and between votes or to law has no carried over, however, amount. No the MOF. Investment "continues the total amount is meet additional provision for this is not regulated by provision for expenditure and expenditures," If unaltered. Spending expenditure with virement and the law. carryover. expenditure from liability already authority lapses at the end virement. Carryover carryover. earmarked revenue may accrued, or expenses of the year or as specified for capital be carried over, already approved, in the appropriations act; expenditures, and for 16 Table 2.4: Public Expenditure Management in OECD Countries - Incentives for Performance Australia France Germany Japan New Zealand Unites Kingdom United States with MOF approval. no authority over five meeting expenditures years. chargeable to that year. Budget Accounting and Cash with accrual Cash Cash Cash Accrual, Generally Modified cash, moving Cash based, with Reporting elements. Accepted Accounting to accrual. some accrual. Comprehensive budget Principles documents and agency annual reports. Audit Financial and Inspecteur des Bundesfinanzhof Board of Audit Controller and Auditor National Audit Office General performance audits by Finances internal performs financial and oversees audits and General focuses on. does financial and Accounting Office the Australian National audit. Court des efficiency audits and reports to Cabinet financial audit, control efficiency audits. performs audits, Audit Office, reporting Comptes external reports to parliament. which passes on system audit. Regular Reports published, and reports congress. to Parliament. audit, reported to Relies on pre-audits by report to _General performance audits, reported to parliament. At state and local Parliament. auditors in Executive Bureau Reports to parliament. levels, most audits administration. does the actual performed by auditing. private firms. Key Organizational Cut in # of ministries Deconcentration to N/A. No new agencies Separation of "Next Step" agencies Privatization, Reforms from 28 to 18 in 1987. regional unless an old one is implementing agencies Privatization. deregulation. authorities. abolished. from ministries. Privatization. Market-type Instruments Increased user charging; Public service N/A. N/A. Contract-based chief Contract based chief Benchmarking; key agencies have charter. executive, state secretary; executives, Citizen's contracting out. published client contracting out. charter; contracting charters. out. Main Legislation/ Financial Management N/A. N/A. N/A. State services Act, Public White papers: Chief Financial Documents for Public Improvement Program. Finance Act (1989) Financial Management Officer Act; Sector Reforms Financial responsibility Act Initiative Next Steps; Government (1994). Initiative; Citizen's Performance and Charter. Result Act. Sources: OECD 1993, 1995, Garamfalvi and Allan, 1997, World Bank, 17 Global Budgetary Targets A Medium-Term Approach to Budgeting Since the later 1970s and early 1980s most Global targets, whatever their form, are OECD countries have adopted global budget translated into annual budgets via their targets in response to growing awareness of interaction with some form of baseline fiscal problems arising from lack of fiscal estimates of the cost of government policy over restraint. Many countries express these targets the medium-term. In OECD countries in the on a multi-year time horizon. Targets are 1960s and 1970s, multi-year budgets were used usually either ratios (of expenditure, budget as planning tools, sometimes in the hands of balance or debt to a national aggregate such as planning agencies with no responsibility for GDP); a rate of growth for spending (real or fiscal management. The total cost of the nominal); or an absolute value for a target policies in the plan became the "global target," variable (such as cash limits on spending or the which was rarely constrained by a realistic size of the deficit). Some countries, such as the assessment of what was affordable in Netherlands, combined deficit targets with aggregate. The plans and targets raised limits on Government's share of GDP. expectations which could rarely be met, and which led to many of the dysfunctional Global targets were questioned in the 1990s budgeting practices referred to above. because they were thought to make the budget more pro-cyclical. Concerns were also raised Beginning in the 1980s, more and more OECD that overzealous commitment to targets would countries adopted a medium-term expenditure lead to across-the-board expenditure cuts that framework (MTEF) in some form or another budgetary reforms were designed to avoid. (Annex 6). A MTEF distinguishes between the Finally, targets often invited creative medium-term costs of existing policies and a bookkeeping on the part of government: in the medium-term perspective on the level of drive to meet the Maastrich criteria, several sustainable spending. The costs of ongoing countries have performed such tricks to reach policy are determined by multi-year cost the deficit target of 3 percent. One of the most forecasts based on a detailed breakdown of creative ones came from a government that sold expenditures for individual policies. a state enterprise to a state bank and counted Macroeconomic conditions, revenue the revenues in its budget. projections, and the sustainable fiscal deficit determine the sustainable spending level. The More recent approaches such as New Zealand's difference between the sustainable spending Financial Responsibility Act and Australia's level and the costs of ongoing projects is Budget Honesty Act define global targets over available for new policies. If this difference is the medium-term, and in terms that allow for too small, governments have to take decisions the business cycle and external shocks. The to drop existing policies, to raise additional strength of these approaches is the transparency revenues, or to refrain from starting new associated with the regular publication of the policies. medium-term fiscal implications of the government's policies. The MTEF provides policy-makers with the information needed to make decisions on new policies that could be accommodated within the Actual OECD country performance in relation to targets has expected resource envelope. The medium-term varied substantially due to fiscal shocks (e.g. German reunification), economic recession or change in political estimates are linked explicitly to annual direction. However, fiscal balances in the OECD did improve budgeting so that annual budget limits are in the decade after 1979, only to deteriorate in the economic firmly based on the medium-term implications slowdown of the early 1990s. More recently, the European Union members have seen marked improvements in their of government policy. iscal balances after adoption of the Maastricht treaty. This treaty restricts membership of a single currency to countries with fiscal deficits below 3 percent of GDP. 18 Australia's experience with the MTEF shows determine what policies were affordable. its potential power to regain fiscal discipline. Together with better incentives for public sector (Annex 7). Introduced in the early eighties, the organization to spend money efficiently (see early projections showed an unsustainable trend below), this allowed policy-makers to in expenditures to finance existing policies. disengage from the day-to-day operations of These projections forced policy-makers to government and to focus on policy. make the choices necessary to reduce expenditure projections down to realistic Over the 1980s and 1990s ministers medium-term targets (Figure 2.1). The forward increasingly took control over strategic decision estimates of the costs of ongoing policy came making and disengaged themselves from day- down from an unsustainable 15 percent in 1984 to-day operational issues. Policy-makers to virtually nil in 1990. Unlike earlier across- moved away from the sequential policy making outside the budget process, which had been an F Etimate 2. usrai B eand Fowad important source of macro instability and poor Estimates, Cumulative Real Growth performance at the strategic and operational levels in the 1970s. Cabinets, often via a powerful sub-committee of key ministers, now focus on the trade-off between priority sectors Ai and programs, within the macro framework discussed above. Governments of all OECD ' \ countries enforce-at least in . .. . ..Nprinciple-collective decisions within the DNS$ executive. Line ministries bought into the new budget processes because they gained much autonomy in making the policy trade-off within their 8M 86 3R 9 91 Q 93 94 9 % portfolio, as long as they stayed within their budget limits. They no longer needed to negotiate with the Finance Ministry for the-board expenditure cuts, the expenditure additional funding, but could concentrate on reductions were driven by decisions about designing new policies. And they had an dropping specific policies. Recently, several incentive to reconsider existing policies, as developing countries have adopted a MTEF, savings could be used for new policies. including South Africa (Box 2.2). Evaluation. Policy discussions were supported Strategic Policy Making at the Center of by improved policy evaluation. The increasing Government interest in outcomes has seen a resurgence of comprehensive approaches to evaluation, the In a medium-term approach, the challenge for results of which fed into the policy discussion. policy-makers is to reconcile what is affordable Canada has had an evaluation system in place with what is required. What is affordable for many years. Australia has made evaluation comes "top down" whereas what is required an integral part of its budgeting process (Annex comes "bottom up." Policy-makers need to 14). Australia's approach is of particular reconcile the two by considering policy interest because of its attention to the incentives decisions together with their spending for evaluation and the associated question of implications, rather than looking at policy and how to link evaluation results into the budget. spending in isolation. Budgeting must In the Netherlands the reconsideration process, therefore become focused on policy. The which involves annual policy reviews, has been MTEF gave policy-makers the tools to 19 Box 2.2: Excerpts from the 1997 Medium -Term Budget Policy Statement in South Africa Relationship with the budget. Government's spending plans for the next three years will be published in March 1998. These plans will give substance to Governments reconstruction and development commitments, within an overall level of spending that the nation can afford. These are the most important choices any government can make. The Budget must reflect Government's social and economic priorities, and its expenditure plans define the nature and scale of the Government's ambitions for the nation. What is the Budget Policy Statement? This Medium-Term Budget Policy Statement sets out the policy framework for the coming budget. It describes Government's goals and objectives. It explains the economic environment within which those objectives are being addressed, and projects the total level of resources that will be available. The Policy Statement analyzes the trade-offs and choices that the nation confronts in addressing its reconstruction and development priorities. The Medium -Term Budget Policy Statement is an important step forward in the budget process. In keeping with our commitment to open, transparent and cooperative policy-making, it invites the nation to share with Government the important choices that must be made. The MTEF Medium-term budgets. The key features of the new medium-term budgeting system are: * publication of three-year forward estimates on Budget Day, consistent with Government's policy priorities and commitments; * detailed analysis of the policy implications of budget projections; * cooperative teams, composed of national and provincial treasuries and line departments, analyzing key sectors and reporting to Cabinet and Executive Councils; * quantified, analyzed policy options presented to political office-bearers for decision; and * the publication of a Medium -Term Budget Policy Statement, to enable Parliament and the institutions of civil society to participate meaningfully in the debate. Rolling budgets. The MTEF initiates a process of rolling three-year budgets. The MTEF projections published on Budget Day will be considered again in the course of 1998, and revised according to new information and policy priorities. The three-year allocations will represent the starting point for that process, and departments will therefore have agreed spending trajectories within which to plan. Departments will be expected in future to frame their policy proposals within their three-year allocations. The introduction of resource-based planning represents a significant change in the planning environment and will initiate a major process of programme reprioritization and redesign within spending departments. Public debate of plans. As in previous years, Parliament will be asked to vote only on the Budget allocations for the coming year, and not on all three years of the spending projections. But the detailed three-year spending plans will provide an opportunity for Parliament, the National Economic Development and Labour Council (NEDLAC) and civil society to evaluate the Budget allocations for the year immediately ahead in the context of the medium-term evolution of Government's expenditure priorities. Parliament will be invited to debate the expenditure plans, and to ensure that they reflect national goals and priorities. This process therefore provides new opportunities for all stakeholders to analyze and discuss the expenditure projections and to ensure that alternative views are fully taken into account in framing the subsequent MTEF. The MTEF and budget reform Overhaul of the budget process. The publication of the Medium-term Budget Policy Statement, and the publication in March of three-year budget projections, are first steps in a wider overhaul of the budgetary process, emphasizing transparency, output-driven programme budgeting and political prioritization. The MTEF provides the bridge between the technical preparation of budgets and the need to reflect political priorities in expenditure plans. Further steps in budget reform will be taken in 1998. These will include a transition to greater devolution of managerial autonomy, within a framework of improved incentives and greater accountability, accompanied by reforms of financial management. Provincial and local governments. The policy goals of Government will be reflected not only in the national budget, but also in the budgets of provincial and local governments. Provincial and local governments will receive their equitable share of nationally raised revenues as well as other transfers. However, they have the responsibility of developing their own budgets, within expenditure allocations consistent with the nation's policy priorities. The national government does not control the details of these budgets, but can influence them indirectly through agreed policies and framework legislation setting norms and standards. The introduction of three-year budgets and their consolidation into resource envelopes for the major provincial services is an important step in the evolution of the institutional framework for intergovernmental policymaking and budget planning. The intergovernmental forums of the spending departments will, for the first time, have expenditure projections within which to develop and refine the norms and standards for service delivery. Role of Accounting Officers. Once Parliament and provincial legislatures have approved their budget proposals, departments must adjust their expenditure to ensure that they stay within their budgets. It is the responsibility of Accounting Officers, appointed by political office-bearers, to ensure that allocations are applied to their intended purposes and spending limits are respected. This principle will be strictly enforced by the proposed Treasury Control Bill which is due to be introduced in Parliament next year, replacing the ten Exchequer Acts which govern provincial and national financial management. Source: MOF. 20 of representative government. A system with Box 2.3: The Netherlands Reconsideration significant distributed power such as the United States emphasizes processes for negotiation .between executive and legislature. Others may The Reconsideration Procedure has been the main betwe. eecuand legislatroe. es instrument of policy review in the Netherlands. It was require mechanisms for compromise because introduced in 1981, and reviews existing policies in the they have minority or coalition executives. The public sector with the aim to develop more cost effective degree to which a legislature changes budget alternatives or to abolish the policy. The Cabinet proposals of the executive varies widely in the decides policy areas for reconsideration. The OECD. In the US, Congress makes many departments responsible for the areas of review need to significant changes, whereas in the UK come up with savings options, including one that would Parliament invariably approves what the reduce overall funding by 20 percent. The Sectoral government proposes, because the proposal is department can use outside reviewers in the team to that of the ruling majority. However, develop policy alternatives. The reports are made public legislatures take an increasingly active role in by means of submission to Parliament. Savings bud proposals are reviewed in the normal budget cycle, but getary policy. reconsideration can be initiated at any point in time. The reconsideration procedure is guided by a small Improving Incentives for Performance ministerial commission led by the Prime Minister, the Vice-Premiers, and the Minister of Finance. MOF The focus on strategic decisions at the center of provides advice and administrative support for the government went hand in hand with major, and reconsideration. sometimes radical, reforms in the incentive structure for public sector units, broadly aiming at increased accountability for performance. in place since 1981 and has played an important The reforms granted more autonomy to government agencies and managers in role in budgetary control (Box 2.3). impementngpocies ard roles an implementing policies, clarified roles and responsibilities of organizations and Transparency. OECD Governments differ on repniltes o ogaztos ad Tranparecy.OECDGovrnmets iffe on government employees, specified the required how systematic and transparent their strategic permen impoed mniino th is decision making process is. Some countries prepare and publish an economic and social performance. strategy which sets out medium-term policy Increased Autonomy objectives expressed in terms of the desired social and economic impacts of policies, their Managers in line ministries and agencies expected fiscal costs and benefits, and obtained increasing flexibility in using money allocation of implementation responsibility to to achieve policy objectives. Detailed central ministers and agencies. Others are less miniter andagecies Oters re ess controls on line items or expenditure norms thorough. All,-however, have some process for cotlsnlietmsrexnduenrs thooug. Al,howver hae omeproessf. have been replaced by more aggregate controls, announcing their policy intentions and their hv enrpae ymr grgt otos th with greater flexibility at local levels to achieve relationship to budgets, and most use the budget agreed objectives within more global speech. Openness and transparency in decision expenditure allocations. To this end, many making increases the incentives for policy- OECD countries relaxed rules of virement and makers to deliver upon their promises, or pay a . carryover. The medium-term approach to political price if they fail to do so. budgeting gave line agencies greater predictability of funding, which allowed better Budgts i OED contris ae wiely planning and allocation by public sector published-more and more often on the Internet. managers. At the same time, aggregate The role ofthe legislature. Decision processes spending limits were strictly enforced by in OECD countries are influenced by the nature improved treasury management systems. 21 Box 2.4: Examples of Output Definitions and Costs in New Zealand Provider Police Output title Custodial services Description Provision of jailing services at police stations and courts for people under arrest or the subject of couri orders. It includes escorting services to convey remand and sentenced prisoners to penal institutions Costs $NZ 15,503,000 Provider The Treasury Output Title Policy Advice: Taxation Output Provision of advice on tax policy, including advice on international taxation, personal taxation Description indirect taxes, business taxation, tax systems, and compliance Costs International Tax NZS 1,286,000 Personal tax NZ$ 562,000 Indirect tax NZ$ 201,000 Business tax NZ$ 2,171,000 Tax system NZ$ 468,000 Ministerial services NZ$ 271,000 Source Purchase agreements between the Ministry of Police and the Commissioner of Police and the Minister of Finance and the Treasury; Quoted in Scott (1996). Output budgeting. Most OECD governments requires that five-year strategic plans linked to still define budgets and control in terms of line outcomes be developed by all federal programs items or programs. In contrast, New Zealand by end 1997 (Box 2.5). By 1999, every agency and the United Kingdom seek to hold is required to draft detailed agency performance organizations accountable for results in terms plans, and integrate these plans into their of the outputs they produce. While these operational and budgetary procedures. The countries acknowledge that outcomes are Chief Financial Officer act requires that each central, they reason that it is not possible to agency be externally audited on effectiveness hold agencies accountable for ultimate impacts, and efficiency. because many other factors affect these outcomes, most of them outside the agencies' Measuring Performance. The slow progress in control (Box 2.4). developing reliable and monitorable outcome indicators confirms the inherent difficulties in New Zealand went furthest in this approach, this area (Box 2.6). Moreover, measuring and abolished input budgeting altogether, to be performance requires not only measuring replaced by output budgets and appropriation outputs, but also costs. Cost information estimates. Within the limits of the collected on an output basis requires revising appropriations approved by parliament, the chart of accounts, revising reporting formats departmental managers have much freedom to and (ideally) adopting accrual accounting. allocate resources the best way possible to Accrual accounting monitors the full cost of achieve the budgeted outputs. government activities, in contrast to the traditional cash accounting which registers only Other countries without an output-based expenditures, but neglects depreciation, budget, have aspects of performance changes in asset position and accrued measurement in their accountability regimes. expenditures. In Australia, resources are allocated by programs. Budget proposals for programs Personnel management. Countries such as include a variety of performance indicators that New Zealand and the UK have extended establish efficiency and effectiveness. These indicators are regularly reported on and orance contract beod athe . organizational level, and contract performance evaluated at regular intervals according to the directly with the permanent secretary or chief evaluation plan. The United States' Government executive. In France, a program for concluding Performance and Results Act, passed in 1993, 22 Bx because the internal organizational culture is Box 2.5: Key Provisions of the Government bureaucratic rather than managerial.4 Some Performance and Result Act of 1993 countries moved away from the strict limits on Agencies must submit to Congress, OMB, and President: personnel numbers and salaries. In the UK and Strategic Plans that cover at least five years and is updated at Australia, for instance, departmental personnel least every three years. First Plan due by September 30, 1997. limits have been replaced by overall limits on Requirements: running costs. At the same time, civil service * include a mission statement career planning and training was enhanced for * include general goals and objectives current and future managers. * describe how goals and objectives are to be achieved * describe how the general goals relate to annual performance goals Redesigning Organizations * identify key external factors that could affect performance Organizational reforms in some OECD * describe program evaluation countries have emphasized separation of functions into different organizations. These Annual performance plans, first plan due to Congress at the reforms promoted clarity and consistency of FY99 budget request. Requirements: * establish objective, quantifiable performance goals; mission and removed conflicts of interest, such * describe operational processes and resources needed to as a conflict between responsibility for the achieve the goals provision of service and the provision of policy * establish performance indicators advice on the same service. Examples of these * provide a basis for comparing program results to organizational reforms include: performance goals * describe the means to be used to verify and validate resu Its* Sweden, in common with other Nordic results countries, has long separated Ministries Annual performance report, first report due to by March 31, from agencies with specific purposes. 2000. Requirements: * The United Kingdom in its Next Steps * describe the performance indicators, actual results, and how they compare to performance goals * review the success in achieving the performance goals of the civil service into Executive Agencies and, if needed, explain why they have not been met, and charged with specific delivery functions. what should be done * New Zealand has restructured most core * describe any waiver of administrative requirement [an government departments to separate policy exception from legal requirements on an agency granted by Congress] and their effectiveness from operations, purchaser from provider, * include summary evaluation findings regulator from operator and inspection * show results for the preceding year from delivery. * The development of Canada's Special Source: National Partnership for Reinventing Government. Operating Agencies, on the other hand, has been a limited initiative affecting only a performance agreements with managers of few of the government's delivery functions. certain well-defined activities has been in place * Similarly, in France there has been limited for many years. Denmark has been use of special administrative agencies implementing multi-year agreements linking outside the basic framework of the budgets and performance. However, several Ministries. countries which have experimented with detailed goal-setting for agencies have stopped OECD experience suggests that reorganizing short of setting individual goals in terms of government can yield large efficiency and outputs or performance, often because this does effectiveness gains, by establishing clear not fit well with the country's culture, or purpose and task; explicit performance expecta- See for example PUMA 1995c pp 176-179 23 Box 2.6: Outcomes, Outputs, Programs, Inputs Increased performance orientation required improved performance definition and measurement. Performance has various dimensions, the usefulness of which depends on their purpose. For policies, and therefore for policy-making budget units, outcomes are the most relevant indicators; for service delivery units outputs are usually more appropriate. Outcomes indicate the ultimate goals of Government strategies and policies. For example, education policies could aim for higher productivity and income for people; health policies could aim for higher life expectancy. However, productivity of people and life expectancy of people depends on much more than education and health policies alone. Policy research can narrow down what effect these policies have on the outcomes, but often not eliminate uncertainty on policy outcomes. The term program is used to describe all the policies, which a government has put in place to achieve a desired outcome. Outputs specify the operational results of departments or other agents of government. For example, education could target 9-year compulsory education, and health policies could aim for full immunization of 5 year olds. These indicators are more specific, they often require quality standards to link the outputs with outcomes. A quality standard for education could be the percentage literacy and numeracy of 12 year olds; and for immunization the range of diseases inoculated for. Significant policy outcomes may be the product of several outputs together with direct budgetary payments and other policy measures such as regulation. Input indicators show resources used in achieving outputs and outcomes. Traditionally, government accounts only registered cash outlays, but increasingly full costs arc measured by including accrual elements in the government's accounting system. Other input indicators relate to actions taken to achieve certain goals such as the purchase of textbooks, the construction of hospitals and the hiring of teachers and doctors. tions; the authority to perform the task and the extent to which they want to use them. pursue the purpose; and which are accountable Besides corporatizing or privatizing market for the use of the authority provided. activities, OECD countries have used a range of devices to create or simulate markets for more Market Incentives core government activities. These include funder-provider models such as used in the UK Improvements in the internal incentives for and New Zealand for health services or in New performance were reinforced by increased Zealand also for science activities and some market incentives for the public sector. education; or voucher systems such as employed in New Zealand and the United Privatization. Over the last two decades, many States for low-income housing assistance. services that were traditionally provided by the public sector were privatized in OECD OECD countries have subjected internal countries. This trend started with utilities, but services to market disciplines by introducing has expanded to infrastructure and user charging for internally provided services telecommunication services as well. Besides a or exposing internal services to competition. In general backlash against government, the New Zealand most central services monopolies privatization was driven by changes in have been eliminated, and the users of these technology that allowed for private provision, services are funded directly rather than funding and by the development of more sophisticated the service. Most other countries however still financial markets that could better finance retain a range of internal service monopolies. large, complex and long-term projects. Contracting out. PUMA reports that experience Competition. The most direct external incentive in the US and the UK with contracting out is to expose government activities directly to indicates substantial savings, but experience competition. Differences among OECD with contracting out relatively simple activities countries' reform strategies are perhaps greatest suggests several essential conditions for when it comes to developing competition. achieving real benefits, notably: Some countries have placed systematic introduction of market elements at the heart of their reforms. Other countries are still studying 'PUMA 1995c p 45 24 * the existence of a competitive market * Service standards. The Citizens' Charter amongst suppliers; introduced by the United Kingdom has * open, verifiable procurement procedures; attracted widespread interest in other and countries. Several countries (Belgium, Italy * solid skills in contract management. and Canada) have instituted specific public declarations of departmental standards of These pre-conditions set practical limits to service. Others are considering them. In contracting out more complex activities in some France, a report7 currently before Ministers countries; at least in the shorter term. And, as is speaks of "putting citizens at the center of the case for all market-type mechanisms, it is public service." evident that potential improvements in * Benchmarking. comparing the performance performance can be dissipated if of an organization with similar ones implementation issues are not given sufficient elsewhere, or in the private sector, has been attention.6 used as a tool to increase competitive pressures on public sector organizations. The case for contracting out is strongest where * User funding. Rather than funding the the good or service required can be specified public sector organization, its users can be with a fair amount of precision, requires little funded. Vouchers are used in some specific capital to produce, and has a strong and countries' education system, funding the competitive private market. Examples are school which parents choose for their computers, transport, cleaning services, children, rather than the school directly. A building maintenance, minor capital works, national health insurance scheme that printing and stationery and office allows choice of provider works in similar accommodation. Some of these conditions may ways. also apply to services such as legal advice and consultancy. Efficient choice between internal The emphasis on responsiveness to citizens and external provision also requires the needs to be reconciled with formal organization's accounting systems to provide organizational accountability upwards. Setting the necessary cost information to set internal standards of service implies that departments prices. will be accountable for achieving them. This in turn logically implies that budgetary decisions The United States has been focusing efforts on will take account of the outputs to which improving service delivery at central level departments are committed. Further, a balance through its National Performance Review has to be achieved between client (NPR) aggressively led by Vice President Gore. responsiveness and client capture. The public The NPR has led to a dramatic reform in service is not seen to be directly accountable to procurement rules with much greater priority its clients. Accountability to the public is being given to value for money. indirect through the government and the Parliament., Competition surrogates. Where services are not contestable, as in the case of most core The Changing Role of MOF' government services, countries have experimented with competition surrogates. The role of the budget office in OECD Such tools are designed to increase countries has changed significantly over the responsiveness of the public sector organization and include: 'Government of France 1995 PUMA 1995c p 61 This section draws heavily from Allan Schick, The changing role of the budget office, OECD Document No. GD(97)109, 'OECD 1995 p 49 Paris 1997. 25 recent reforms. The traditional role of the MOFs in several OECD countries have taken a budget office (usually MOF, although some leading role in public expenditure management OECD countries have budget offices separate reforms. However, reforms have only worked from the treasury function) was to function as a in countries in which a broad commitment to central command and control post, specifying reforms existed. the items of expenditure, negotiating budget proposals with line ministries to control overall Conclusions expenditures, monitoring compliance with regulation, and ensuring that inputs were used The OECD experience suggests that key as agreed in the budget. elements of a performing public sector include: (i) hard budget constraints based on a multi- This role has changed in those countries that year perspective on the budget and enforced by have undergone significant reforms. Managing internalized discipline of budgetary units (ii) the budget process is still the central role, but results oriented institutional arrangements for the way this role is operated differs from the decision making, resource allocation, traditional role: budgeting, personnel management; and (iii) organizations which have clarity of purpose and * MOF is usually in charge of the multi-year task, clearly specified performance forecasts: it either makes them itself (as in expectations, authority to perform the task and Australia) or manages the process by which commensurate accountability mechanisms. line ministries make the forecasts. In the latter case, MOF provides the key * Reforms are still ongoing, and some economic data driving the multi-year countries have only just begun. It is forecasts, and guides ministries on how to therefore too early to tell whether the assemble the forecasts. reforms have been successful. However, the growth of government slowed in most * MOF usually manages the process of OECD countries, and was even reversed in strategic policy choice. This no longer some. Deficits improved throughout the implies making allocation decisions during eighties, in particular in the forerunners of budget negotiations. Rather it means reforms, New Zealand and Australia, facilitating the process by which line although a relapse occurred in the recession ministries make trade-offs within the of the early 1990s. In the USA, in part due overall expenditure limit. MOF usually to an unusually long upswing in the sets guidelines and procedures by which economy, the large deficits of the 1980s line ministries propose new policies and have been eliminated altogether. But review existing policies. perhaps the most significant result of public sector reforms is a reassessment of the role * MOF's role in budget implementation has of the state. At the end of the 1970s, become one of monitoring and informing government was seen as the problem rather rather than control. Aggregate spending than the solution. Now, the emerging limits are still rigorously enforced, but line consensus is that the state has a significant ministries have gained significant leeway in role to play in the economy and that detailed spending decisions. In return, capable states can do more and with better MOF's task is to ensure that resources are results. used productively. To this purpose, MOF has taken the lead in revising budgeting, accounting and reporting formats to match new concepts of performance and to ensure that this information is regularly reported. 26 CHAPTER 3 TOWARDS A MORE COMPREHENSIVE BUDGET The rapidly growing extra-budgetary and off budget funds-now almost half of all government resources -undermine fiscal discipline, hamper funding of the government's priorities and give rise to wasteful spending. However, these funds have also served well to finance local initiatives outside the rigid planning system. China needs to better integrate EBFs into budgetary management, but should carefully balance the incentives for revenue mobilization, service delivery, and expenditure control. Extra-Budgetary Funds aside from the budgetary process under decentralized management and allocation by China's extra-budgetary funds (EBFs) comprise different agencies and organizations. surtaxes, levies and user charges accruing to government and administrative units.1 They During the 1980s EBFs grew rapidly with were originally created in the 1950s to set aside decentralizing reforms of the economy. Official small amounts of funds for the discretionary use statistics show that they grew from 2.6 percent of of local governments, administrative agencies GDP in 1978 to over 4 percent of GDP in 1995. and enterprises. Examples of these early EBFs A 1996 audit found that EBFs totaled more than are a surcharge on the agricultural tax that was Y380 billion, or more than 6 percent of GDP, and used for rural social expenditures, and a far exceeding the officially reported amount.2 surcharge on the Industrial and Commercial Tax Subsequent audits unearthed more unauthorized that was used to finance urban maintenance and funds and fees at all levels of government. The construction. Rental incomes from public best estimates now peg EBFs at 8-10 percent of housing, users' fees, surcharges for public GDP, but run as high as 15 percent (Box 3.1). utilities were also counted as EBFs. All EBFs Local governments rely to a much larger extent were earmarked for specific uses and were set on EBFs than the central government. Local Box 3.1: "Would you like that with fees?" McDonald's Restaurants in Beijing pay 31 fees on average, of which only 14 are legal, all this beyond the normal taxes due. With those fees, McDonald's supports not only the normal Beijing Municipal services, but also air shelter repairs, river cleaning, public festival decoration and communist party propaganda. The restaurant also pays for not having security guards-which is still cheaper than having one. The illegal fees and fines and various apportioned charges that the many central, municipal, county, and township authorities impose on enterprises and individuals increasingly affect normal business operation, so much so that it is now seen as a deterrent for foreign investors. This has prodded the State Council to issue a circular on July 7, 1997: "On Abolishing Illegal Fees and Fines and Various Apportioned Charges for Enterprises." The circular orders that any illegal charges, fines, fund collections, foundation projects and other apportioned charges to the enterprises must be abolished. Abolishing the fees is easier said than done, however, as repeatedly failed attempts to limit the taxing of farmers have shown. The key reason for failure to resolve the illegal fees is that, notwithstanding some excesses, many of the illegal fees pay for useful services, or pay the wages for civil servants that cannot be fired. Source: China Financial Times, Asian Wall Street Journal, State Council 2 MOF, SDPC, the State General Audit Agency, PBC and the Until 1993, EBFs included enterprise funds-retained profits, Ministry of Supervision conducted the audit. See China depreciation, technical renovation and welfare funds, which were Securities January 29, 1997. at their peak as much as 80 percent of the total EB3Fs. The 3 Officials from the MOF tend to cite the lower figures, while current definition of EBFs includes only "fiscal" EBFs-those those from the State Administration of Taxation cite the higher accruing to governments and NPOs. ones. 27 used EBFs as de facto taxing mechanisms, a outright borrowing from banks. Revenues practice tolerated and even condoned by the from some state asset sales such as land central authorities., leases have also been kept off budget., Although Government considers EBFs to be * Social Security Funds. China's social fiscal resources, they are managed separate from security funds are administered off budget at budgetary funds: local level. Provincial pooling of pension * The funds are allocated outside the budget and unemployment funds has started, but process. Extra-budgetary allocation takes most funds are administered at municipal place when they become available. levels. At present, these funds run a slight * The funds are reported outside the budgetary surplus, but at current levels of benefit, the reporting system. Reports on EBFs are pension funds are seriously underfunded. infrequent and final accounts are only Unemployment funds are also likely to run available more than a year after the budget deficits, once SOE reforms will accelerate, year closes. and hidden unemployment becomes visible. * The funds are managed through bank Social security funds are reported in official accounts separate from budgetary funds. statistics, but are not considered fiscal in China. * The funds have a separate accounting system and accounting classification. Tax Expenditures. Even less visible are China's numerous tax expenditures. A Other offbudget funds. Besides EBFs, China has Chn' .ueosta xedtrs Oumerusother of f-budget funds. e B s C a thse variety of industries receive tax exemption or numerous other off-budget funds. Most of these txrdcin npriua ihtc are not considered fiscal funds, and not mecluded txrdcin npriua ihtc are ote ficnsiderednfscalnds,e an not included industries, foreign invested enterprises, and in the fiscal accounts, and some are not reported enep.e nseil cnmczns hs at a I.Off-udgt fnds nclde:enterprises in special economic zones. These deviations from normal tax policy have effects similar to expenditures: granting tax NP S anage"self-raised Fundf.uLocans a) exemptions for, say, technology, means that NPOs manage "self-raised funds, (SRFs)", lesmnycnbsptfoedain.n compisin Tonshi an Vilage less money can be spent for education. In prisg (Toshi and ilagce the same vein, too much tax expenditures Enterprises' (TVEs) profits and remittances, unemethrvneba,admycus manaemet fes ollctedby POs retal undermine the revenue base, and may cause management fees collected by NPOs, rental budget deficits. Some estimate the total "uncomed from"(tongeo) collect s e d tm amount of tax expenditures to be Y100-150 "uified levy (tghousehold cactd frm billion in 1995.7 Others have estimated that farmthousehlds fra ni acinglrural in 1994 tax reductions and exemptions to edctin miii riig odbidn, foreigners added to over 1 percent of GDP8. family planning and social welfare. In 1993- 1994 SRF amounted to about 20 percent of total revenues accruing to township level governments.5 Other SRFs include (illegal) local government bonds, "contributions" and For example at the township level off-budgetary resources are explcity rcognzedforfinncin goernent-inc thir The State Land Bureau estimates that over the past few years explicitly recognized for financing government-since their revenues from sales of land leases have totaled more than 25 founding in 1986, township finance bureaus have been required billion Yuan nationwide, only a fraction of which were reported as to report receipts and expenditures of "three types of funds": fiscal revenues. budgetary, extra-budgetary, and SRFs. A State Council decree in 1991 explicitly permitted township and governments to impose a 7 Source: mission interviews. it is not clear whether the 1996 audit fee on TVEs. to finance agricultural subsidies, called the "using captured these revenues. industry to aid agriculture" fee. s IMF, Article IV consultations, annex to the report on Recent Economic Developments. The most important one is the 5. Ministry of Finance. exemption from import duties on import of capital goods by foreign funded enterprises. 28 * Arrears: Finally, regular reports on arrears the budget from the budget allocation process appear in Chinese newspapers: teachers go and even from budgetary funds management. without pay, and grain bureaus do not pay Examples are the Statutory Bodies in farmers for their grain. China's cash- Australia, Crown Agencies in New Zealand, based system does not capture these the rechtsfaehige Anstalten in Germany, and arrears, which make the budget deficit on the Etablissements Publics such as museums in an accrual basis larger than that on a cash France. These public agencies are established basis. under separate laws, and have budgeting and financial management rules separate from the An International Perspective on EBFs bde a.Otn hyhv oemn budget law. Often, they have government Practically all OECD countries, developing representatives on their board. The budget countries, and in particular countries in only shows the subsidies to those transition have EBFs (Table 3.1). Some of organizations. Such organizations are not these funds bear similarities to China's EBFs, unlike SOEs, which are usually also treated on but there are fundamental differences. OECD a net basis. "extra-budgetary transactions" comprise all government transactions that operate outside The Russian practice with many industry- the normal budgetary procedure. This specific EBFs comes closest to the Chinese exclusion can be at the funding stage (if situation, showing the common roots of the budgetary revenues are set aside for a budget system. The Korean EBFs are also designated expenditure), at the budgeting stage similar to the Chinese ones. In both Russia (if financing is not fully determined by budget and Korea, however, the number of EBFs is deliberations), or at the funds management far smaller than the 900 types of funds in stage (if EBFs are organized as accounting China. entities with own accounts). Minimizing Disadvantages of EBFs Unlike in China, OECD EBFs need not be EBFs can have advantages for certain types of funded by earmarked or extra-budgetary expenditure (Box 3.2). However, often revenues but can be funded from budgetary disadvantages dominate. Practices that tend to appropriations. Earmarking within the budgetary framework is commonly done miiiethe nti Effe through special accounts, which pool the revenue from one or more sources and is accessible only for specified expenditures.9 * EBFs should be constituted by law and These revenues and expenditures are subject to respond to well-identified criteria, so as to the normal budgeting procedure at the avoid the unnecessary multiplication of execution stage. The difference is their EBAs. In France and in the United States, privileged access to the special accounts. Such among others, law must institute EBFs, earmarked funds protect funding within the whereas in Russia they can be established budgetary allocation, but they do not protect by decree. In Germany and New Zealand, the activity against revisions of the budgetary non-appropriated expenditures may be allocation from year to year. Annex 13 incurred provides an illustrative example of earmarked funding of highway construction in the US. Many OECD countries exclude some self- raised funds of agencies largely dependent on In the United States, the budget includes several earmarked Funds, which are denominated "trust funds". 29 Table 3.1: International Comparison of EBFs USA Russia Germany New Zealand France South Korea Types of Extra- Two federal fiscal 5 social funds. Social security fund. Unappropriated Social security fund. Around 30 EBFs "Special-Budgetary budgetary trusts funds: Technological expenditures. Funds" (e.g. Grain Management fund, Resources. 1) Old age and development fund. Unappropriated Contingent liabilities abolished 1994). Survivors fund 50 industrial funds expenditures. Permanent 2) Disability fund (R&D, Investment, and appropriations for Dependent organizations; Credit subsidies. Economic Distress public debt and ODAC. funds) constitutional Directed state organisms. Implicit tax and credit subsidies. Cost recovery. Revenue Employers and Social fund: employers Social security financed General revenue. Social security financed Special funds financed by earmarked taxes, Source employees and employees through employers and through employers and and transfer from budget special accounts. contributions contributions. employees contributions employees contributions (payroll taxes) Industrial funds: and through budget and through budget percentage of transfers. transfers. production costs or of sales. Rules of EBF By law EBFs: by law or decree. Unappropriated Permanent Established by law, with time limit, under Constitution expenditures authorized appropriations: by law. the authority of a line ministry. Cost recovery: decision by the MOF only under of the levying office. unforeseen and Unappropriated compelling expenditures: maximum Direct loans: MOF circumstances, and 1% of ministerial decision either below DM10m or appropriation with it is a legal obligation. authorization of MOF. Budgeting The off-budget EBFs not included in Social funds prepare Completely integrated Social security prepares Budgeting done by line ministry, and (organ, funds are included budget document. their own budgets. The with budget. its own budget, which is approved by the president. integration with in the budget budget includes only the discussed during the main budget) presentation. Line ministries make flows to and from the budget session. The Unified budget totals are presented with the Unified totals are budget of EBFs. social funds and the parliament can impose an budget document. shown for budget. annual spending limit. expenditure, Heads of agency decide Unified total is presented revenue, and deficit. on level of cost as an annex to the budget recovery public service. document. 30 USA Russia Germany New Zealand France South Korea Treasury Social security funds Industrial funds: Social security has its Permanent Funds kept in a separate Fund kept in separate bank account management. manage their own revenue should be paid own accounts. appropriation: same as treasury account, under authority of line ministries. accounts. but within to line ministries who for normal budget managed by an Treasury can borrow from surplus the Treasury system. distribute it. In practice, items. autonomous agency funds, after negotiating with the line Treasury borrows at large portions are kept (URSAF). ministry. market rate from by enterprises and spent them. or kept in bank accounts. Transactions Only limited EB funds keeps their Budget transfers of , Not applicable. Budget finances deficit of EB funds keep surplus or lend it to with budget financing from surplus, although, for finance social security. social security. treasury; get budget funding or budget funds. general revenue, for industrial funds, part of loan for deficit. unfunded surplus is kept by entitlements. enterprises belonging to the fund. Accounting & No full consolidation No full consolidation No full consolidation Accounting: same as for Separate accounting. No EB Funds use commercial accounting consolidation with budget accounts. possible with budget with budget accounts. budget. Full full consolidation of system. No consolidation. with budget accounts, consolidation. social security accounts with budget accounts. accounts. Reporting and No reporting to No reporting to Same as budgetary No regular reporting to No regular reporting to treasury. Monitoring. Treasury. Treasury. items. treasury. Subject to the same external auditing practices. 31 Reforming China's EBFs only under exceptional events, and for limited amounts. China's public expenditure management is * The extra-budgetary resources should be severely weakened by the large and growing kept under the authority of the Treasury, amount of EBFs and other off-budget possibly with some special procedures to financing. The large amount of EBFs guarantee the priority of funding. This fragments government resources; threatens facilitates cash management, while aggregate fiscal discipline, and reduces assuring the regularity of funding for the incentives for agencies to operate cost- extra-budgetary activities, effectively. * Accounting should allow easy consolidation of EBAs with the budget State Council Document No. 29 (Annex 2) is accounts, and the central budgeting agency an important first step towards a more should have unrestricted and automatic comprehensive budget. The main changes access to the EBAs. If this is not feasible, under Document No. 29 are: at the minimum EBAs' totals should be added to the budget totals to provide a * It extends budgetary management to unified measure of public fiscal activity. thirteen funds that had been treated as * Reporting and auditing requirements "below the line" items in the budget. should be identical to those of the budget These amount to a total of Y1 15 billion, accounts. For example, non-appropriated out of a total of Y3 80 billion. expenditures are fully reported in the New * It increases MOF supervision of the Zealand Government financial statements. allocation and disbursement of funds. * It shifts ownership of revenues collected under EBFS from the collecting unit to Box 3.2: Pros and Cons of EBFs EBAs are mainly used for activities which are to a large extent self-financing but which are not suitable for commercial provision, or whose provision the Government wishes to guarantee. The purpose of EBA accounts is to protect some activities from regular budgetary reviews of financial allocations, and to ensure the resilience of funding over time. The question is why certain expenditures should escape the screening and weighing process of the budget. EBFs can be useful for: * financing certain activities for which the public pays a price and expects a service in return. The budgetary decision process would delink price and service, by determining the price and the funding for the service independently. In this case, extra-budgetary finance is a specific form of earmarking. Social security or road maintenance are examples of such services. However, individual goods, such as electricity or water, are usually not financed through EBAs because they can best be supplied by profit-motivated enterprises. * showing government commitments to a long-term moral or legal obligation, such as war veteran pensions, or the reimbursement of public debt. This gives rise to permanent appropriations which cannot be modified during the budget process, but require a different legislative process and sometimes constitutional amendments. * responding quickly to expenditure needs arising from unforeseen events. Unforeseen events should be distinguished from risky activities such as government guarantees. The budget should account for the risks it assumes as a contingent liability, and budget for contingent expenditures. The adverse effects of EBAs include: * weaker macroeconomic management, by barring a comprehensive assessment of all expenditures and revenue, and hindering expenditure control. This it turn impairs the ability of the fiscal and monetary authorities to act consistently and incisively to ensure macroeconomic stability. * increased rigidity of resource allocation, which affects policymakers' ability to fund strategic priorities. * lower efficiency of service delivery, as they encourage overspending in those areas where (earmarked) revenue is plenty. Also, the lack of a comprehensive budgetary framework for EBAs can lead to the duplication of expenditure or taxes. * looser financial control, as the lack of standardized accounting and reporting rules, and the complexity of transactions between thEBFs and the budgetary funds makes financial management intransparent. * inefficient cash management, because idle balances are held in many separate accounts, thereby increasing the borrowing requirement of the government, and its interest payments. EBAs share with earmarked funds the property of increasing the rigidity of public resource allocation, and making it difficult to use the budget to implement strategic policy choices. 32 MOF and local finance departments. The budget, the coordination between planning and comprehensive department is put in charge budget would have to be significantly of managing these funds. strengthened (see Chapter 4). * It announces the intent to convert some "fees" to "taxes." For these reasons, reintegrating EBFs with the * It clarifies and sharply limits the authority budget must be a gradual process with several for setting fees and levies, stipulating that difficult policy choices to make. In line with only the central government can set up State Council Document No. 29, and building "funds", and only central government and on the 1996 audit of EBFs, the authorities provincial governments can levy fees. would need to review all of the hundreds of * It eliminates the category SRFs at the types of legal fees, levies, charges, and township level and below and folds the surcharges and decide whether to continue funds into EBFs, while extending coverage them as EBFs. Some of these funds, starting of EBFs to include other non-budgetary with the illegal ones, will be abolished. EBFs revenues collected by township that are truly taxes, should be fully integrated governments including various voluntary in budgetary management, and treated as contributions, revenues from fiscal loan general revenues. Others, such as surcharges repayments, and interest from fiscal bank for specific goals, could remain outside the accounts. budgetary allocation process as earmarked funds, but would be fully integrated in With these changes, China has taken a decisive collection, treasury management, reporting and step towards consolidating budgetary audit. And finally, some user charges may be management, gaining control over the excluded from budgetary management aggregate spending level, and curbing the altogether, and be used for funding proliferation of funds and fees. In 1997-1998, commercialized NPOs. MOF began a thorough review of existing fees and funds, abolishing some, reducing the rates In the short run, Government could focus on of others, moving some into budgetary improved reporting and funds management of categories and excluding the business incomes EBFs. Over time, emphasis should shift to catgofr s und erlunme aeness iconsolidating the number of EBFs, and to simplify collection. Finally, the hard question In further integrating EBFs, however, the whether Government has a role in the activities authorities face a delicate trade-off between financed by EBFs should be tackled. incentives for service delivery and revenue mobilization by the spending units, and Improvefundsmanagementandreporting improved treasury management and expenditure control. Ignoring these trade-offs * Improving reporting. A first step to may result in a rapid recurrence of off-budget improve reporting could be to allow the funds, as happened in the past. Notably, EBFs collecting spending unit to retain the play an important role in funding local funds, but to require them to present governments. Abolishing them, or reducing sources and uses of EBFs together with the incentives to collect them by exercising too budget proposals and budget reports, even much central control may leave important if the National People's Congress (NPC) services unfunded, or may force local only allocates budgetary funds. During governments to invent new levies and fees. budget implementation, budget units Finally, many EBFs finance investments should be required to report on EBFs as controlled by the State Development Planning frequently as budgetary funds. Over time, Commission (SDPC), or finance the recurrent the reporting system, and the accounting costs for maintaining and operating past and classification of budgetary and EBFs investments. If EBFs are integrated in the could be unified. 33 agents in the collection of most fees and * Consolidate bank accounts and treasury charges at much reduced costs. management. The supervision of remaining EBFs could be further improved Reconsider the role of Government by integrating the budgetary and extra- budgetary bank accounts of budget units. * Reconsider role of Government: In This would have to fit into a more general reviewing EBFs, government would need overhaul of treasury management and to decide which tasks financed by EBFs accounting practices in China (see Chapter are worth continuing, and worth 5), but already there seems to be few continuing as government tasks. Divest reasons for keeping separate budgetary the activities that are no longer considered and extra-budgetary bank accounts for government activities, either to the private expenditures, and for keeping separate sector, or to commercialized SOEs (see disbursement processes. also Chapter 5). Consolidate funds * Reconsider government organizations. The review of EBFs needs to go hand in * Review taxes, levies, fees. Many fees hand with a review of organizations within raised in China are truly taxes, and should government. NPOs that are fully or be treated as such. Examples of this type largely funded by user charges might be include surcharges for rural education better off as more independent units, or levied on urban building construction, and even enterprises. In some cases, this may the education surcharge on the VAT and require increases in user charges to cover other turnover taxes. Revenues from user long-term costs. Even some organizations charges should remain with the collecting that still depend on significant budgetary unit for financing the services for which contributions could be better managed they are levied. Revenues from taxes outside the budgetary framework. The should go into general government "Regulation on Financial Management for revenues and used to finance the broad Non-Profit Making Entities" of 1989 range of government services. would need extension and revision to define a framework for such independent * Consolidate remaining funds. NPOs, including budgeting and financial Consolidating the fees and funds into reporting rules. Until such a revision has fewer items and categories should help taken place, Government could consider simplify the payment structure and reduce treating NPOs as autonomous budgetary the costs of collection and administration. agencies governed by a contract that Some of the surcharges on taxes could be specifies revenue retention arrangements integrated in the tax rate, others that have and expected performance. outlived their use could be abolished and replaced by the increasing general * Reconsider intergovernmental fiscal revenues. Reducing the number of relations. Because EBFs have become a collection units would reduce the major source of local finance, integrating taxpayers' burden of dealing with a them in the budget is likely to require multitude of collectors, and would reduce adjustments in intergovernmental fiscal collection costs. For example, the relations. If extra-budgetary levies are collection costs for the township-unified consolidated with the tax structure, a levy, collected from farm households, reassignment of tax revenues may become absorb as much as 25 percent of revenues necessary. Alternatively, higher levels of collected. Local tax bureaus can act as government could consider earmarked 34 grants for the activities previously reported together with the budget, and China financed with EBFs. should aim for the same. Together with improved reports on EBFs, this would * Unify budgeting procedures. Over time, establish a comprehensive reporting on for those EBFs that are truly government General Government (Figure 3.1). funds, budget procedures for budgetary and EBFs should be merged. Ministries Contingent liabilities. Because contingent and budgetary agencies would have to liabilities could greatly affect China's budget include the sources and uses EBFs in a at some future date, regular review of the size single budget proposal. MOF's of these liabilities is a priority. In OECD organization would have to adapt to the countries, reporting on these liabilities is far unified budgeting procedures. The Budget from comprehensive. New Zealand has the Law already allows for such an integrated fullest treatment of "fiscal risks," including budget. A first step in that direction would contingent liabilities-claims against the be to unify the accounting system for government which will arise if a specific event budgetary and EBFs. occurs-and any event which if it occurred, would affect the revenues or expenditures of A Halfwiay House? the State.'0 Most OECD countries report on For some of the EBFs Government decides to loan guarantees given, either by simply listing integrate further into the budget allocation them, or, as in Germany, by budgeting for the process, the Japanese budget system might be likely default on the loans guaranteed. a useful halfway house (Box 3.3). The Japanese budget separates the general accounts To start reporting on contingent liabilities of from 38 special accounts. The special the budget, the authorities could first make an accounts are financed by general revenues, inventory of possible liabilities that may earmarked taxes, and user charges, and are accrue to the budget. Undoubtedly, those in earmarked for certain uses. They are approved the banking sector and in social security will by parliament as part of the normal budget constitute the bulk of it, but there may be process, and managed throughout the treasury more. China's policymakers would then have accounts at the Bank of Japan. However, the to decide for which liabilities they would have Japanese system maintains some of the to or want to take on responsibilities. For disadvantages of the current situation in that those, the authorities would need to design significant parts of the budget are earmarked, reporting formats and reporting requirements. and the system can therefore only be However, fiscal reports on the public sector transitional. are usually restricted to the public sector deficit. Statistical bureaus, together with the national accounts usually compile reports on How to treat other off-budget funds the complete accounts of the public sector. Social Security Funds. Social security funds lend themselves less for full integration in the Tax Expenditures. The sheer size of China's budgetary framework. However, reporting on tax expenditures and their similarity to these funds could be greatly improved. In expenditures necessitates a fuller reporting most countries, public social security funds are over these funds. Many OECD countries include reports on tax expenditures with the budget (e.g., the USA and New Zealand), or regularly present reports to Parliament (e.g., Underpinning the New Zealand approach has been a shift to a full accrual approach to budgeting. 35 Germany's bi-annual Subventionsbericht). For not only consider whether the tax China, if not the NPC, at least the policy expenditures still support their priorities, but makers should be made aware of the resources also whether the instrument of tax expenditure foregone in giving tax reductions. They can is still appropriate for achieving the objectives, then regularly decide whether the tax or whether others such as subsidies may be expenditures still match their strategic more appropriate. priorities. Over time, the Authorities should Box 3.3: Japan's Budget: General and SpecialAccounts The Japanese budget system might be a useful halfway house for some of theEBFs Government decides to integrate further into the budget allocation process. The Japanese budget separates the general accounts from 38 special accounts, grouped into 8 groups (Table) Box Table: Japan's Special Accounts Type Number Explanation Examples The enterprise 5 To manage specific Mint Bureau, Printing special account government administered Bureau, Postal Service enterprises The insurance 11 (a) Special insurance, (a) welfare insurance, special account reinsurance, and social National insurance, insurance that the private labor insurance; sector does not provide; (b) Postal life insurance, (b) insurance comparable to forest insurance private insurance Special account for 5 for certain public works Port improvements, public investment road improvements, flood control Special account for 9 to manage administrative national schools, administrative business, except insurance hospitals business and public works Special accounts for 2 to manage the loan program industrial investment loans to public corporations for Urban development loan specific purposes Special account for 2 to manage public funds Trust fund bureau fund; fund management established for specific foreign exchange fund purposes The consolidated 4 to consolidate and account National debt fund special for revenues and consolidation fund; account expenditures for specific local allocation and purposes transfer tax The special accounts are established by legislation under specified conditions. These may be instituted when Government carries ou specific projects, manages specific funds, or when it deems it necessary to administer some revenues and expenditures from the genera budget. The special accounts are financed by transfers from the general account, user charges, insurance premiums, and loans from th( fiscal investment and loan program, which captures the postal savings and some social insurance funds. Both the general account and th special account are reviewed and approved by Parliament as part of the normal budget process. Source: MOF Japan 36 Figure 3.1: Definitions of Government Central Government + Local Governments Government + Social Security and EBFs General Government + Nonfinancial State Enterprises + State Banks Public Sector 37 CHAPTER 4 LINKING POLICY, PLANNING AND BUDGETING China's budget cannot yet play the more prominent role it will have in the market economy. Recent budgetary reforms and experiments are promising, but more far-reaching reforms should move China away from its current incremental budget process. MOF must bring a multiyear perspective in the budget, and forge a stronger link between strategic priorities of Government and public expenditures. The Government needs to reorganize its budget process such that policy and spending decisions coincide. It should require more detailed budget proposals, and build up evaluation capacity to ensure the funding of policies that support priorities. Better coordination between MOF and SDPC, and between central and local governments could help China to achieve its policy goals at spending levels compatible with macroeconomic stability. When the Plan was the Government's main tool for longer-term impact on the budget, and on the for implementing its policy priorities, the budget policy outcomes desired by Government. was the most important source for financing the plan. No independent fiscal policy existed. With The macroeconomic framework. China's fiscal the move to the market economy, the budget will policy is anchored in the long term, 5 year, and become China's most prominent policy annual plans. SDPC, with inputs from PBC, instrument, as the control over SOEs and Bank's MOF, MOFTEC, SETC, and other agencies draft diminishes. these plans. The long-term plan includes broad projections on the main economic indicators, and Budgetary policy will, together with monetary spells out areas of reform for the following 15 policy, largely determine macroeconomic years. The five-year plans outline conditions for China. And the budget process macroeconomic projections, including growth will largely decide which Government policy rates and investment rates, and the broad gets funded and which not. In the transition to the financing needs from budget and credit plan to market economy, China's budget should be finance planned investments. MOF compiles the agiled enough to take on new tasks such as five-year fiscal plan, which includes broad providing social safety, while cutting back on old projections on revenues and expenditures. ones like financing. When EBFs are integrated However, the plan has no operational function, in the budget, the budget will gain in importance and is not updated until the next five-year plan. for China's policymakers. But the budget needs retooling, before it can assume larger The annual plan for economic and social responsibilities. development provides the macroeconomic context for annual budget compilation. The State China's Incremental Budget Process Council is in charge of formulating the plan, but SDPC does most of the work. In the fall China has a highly incremental budget process preceding the budget year, SDPC forecasts (Annex 1). The extensive amount of strategic economic growth of the coming year; MOF, planning and the explicit priorities of the State SAT, and Customs project budgetary revenues, Council are insufficiently reflected in budgetary and MOF makes expenditure forecasts for the allocations. The incremental budget process coming year. SDPC, PBC and MOF conduct undermines the Government's aggregate fiscal meetings to discuss economic conditions, and goals, fails to shift expenditures to emerging SDPC and MOF negotiate the budgetary priorities, and dilutes priority spending during the contribution to the investment plan. Future fiscal year (Chapter 1). New and existing recurrent costs of investments are not taken into spending proposals are insufficiently scrutinized account in the budget. Problems with funding 38 budget. Problems with funding operations and level cannot agree on spending. Some ministries maintenance have remained limited because most see only 20 percent of their proposals for recurrent costs for investment projects have been additional expenditures approved at the end of financed from EBFs. the budget process. The low approval rate is partially due to a negotiation strategy: Non-binding fiscal targets. Overall fiscal targets overdemand for funds demonstrates the needs of play a small role in the budget process. The goal the sector, and proposals that are turned down of the current five-year plan is to "basically repeatedly have a higher approval chance in a eliminate" the existing budget deficit.' However, next budget round. despite revenue increases of 10 percent or more (and 8 to 9 percent in real terms), the last two The budget proposals of individual ministries budgets made no progress towards this goal.2 lack sufficient detail to determine the links The target is already downplayed by the between the proposals and the priorities of the authorities: The Finance Minister stated in the Government. A typical budget proposal would March 1997 budget speech that "The deficit in contain summary descriptions of new the central budget is still so large that the expenditures, and a table of expenditures on objective of reducing the deficit set in the Ninth ongoing programs. Running costs for ministries Five-Year Plan will be difficult to achieve." and agencies, split up in "operational costs" and personnel costs, are separately budgeted for. Little focus on strategic priorities. The current Budget proposals do not contain indicators for budget process does not support a strong link expected results. Output and outcome indicators between budgetary policy and strategic priorities. are lacking, although sometimes spending Its central purpose is to allocate the incremental proposals indicate the activity to be financed. revenues over new spending proposals: existing expenditures are hardly discussed, and the The focus on funding rather than policies is process is largely bottom up. A State Council reflected in the annual budget speech. The Circular that starts the budget process includes budget speech lists planned and increased general guidelines for expenditure increases. expenditure over the previous year. Few, if any The circular gives some guidance per sector, but concrete policies are announced in the speech, sectors are usually subdivided over a range of although it does mention some broad priorities line ministries. No line Ministry therefore feels ("support of agriculture, education"). The bulk of responsible for meeting the global or sectoral the text is devoted to spending levels on the targets, and budget requests supersede the limits various budget categories. This contrasts sharply by a large margin. with budget speeches in most OECD countries, which often announce major policy initiatives. Because of the lack of ministry-specific spending limits, much of the budget process is used for No Organizational Breakdown. A fundamental negotiations between spending ministries and the weakness of China's budget is that, apparently, corresponding sectoral department in MOF. At no organizational breakdown of spending is this level, numerous policy decisions are made approved by the National People's Congress on the basis of whether additional funding is (NPC). Although such a breakdown may exist in available, not on whether the proposed policies internal documents of the administration, the meet the Government's priorities. Decisions are absence of firm spending limits per line ministry referred to a higher level only when the working makes the sectoral budget limits nonbinding. The sectoral budget becomes a common pool for It is debatable whether such a goal is appropriate for an economy individual line ministries that operate in the growing with 9 percent per year. However, that is not the subject of this study. 3 In fact, each ministry submits two proposals to different sectoral The budget speech has reported a sharply falling deficit, but this is departments in MOF: one related to the ministries' expenditures the result of the exclusion of interest payments from expenditures. and NGOs, and one related to State enterprise expenditures to the Trade and Finance Department of MOF. 39 Box 4.1: An In-Year Budget Request Two months after the budget passed, a line ministry approached MOF for additional funding for a certain high priority function. To increase its capability in this function, the Ministry asked for (in order of preference): (i) a tax earmarked for this function; (ii) a special fund for this function; (iii) a rule to increase expenditures on this function by more than revenues; (iv) earmark the fines the ministry collects for this function; and (v) a grants system to counties for the specific function. In an OECD country, the MOF's answer to the request could be as follows. First, the budget has just been passed, and the ministry would have to wait for next year's budget round. Second, if the function falls within the Government's priorities, financing should first be sought within savings from the ministry's other, non-priority activities. If budgetary money were still needed, the proposal would have to specify: the concrete activities planned for achieving the Government's strategic goals. Furthermore, the line ministry should specify measurable indicators for success of the policy, and-in some OECD countries-would be required to present an evaluation plan. Even if the final proposal were satisfactory and accepted during the regular budget round, it would rarely happen in an OECD country that the line ministry would obtain earmarked revenues for the function. Rather, it would be funded from general revenues. sector, with little power on MOF's side to resist of spending under either Plan or Budget is requests for additional spending (Box 4.1). A undermined., Until there is a more effective link characteristic of well-performing budget systems between investment decisions and the associated in OECD countries is that the budget approved recurrent spending-with both driven by policy by the legislature (National People's Congress priorities-it will be difficult to assess what (NPC) in the case of China) sets firm limits on services are being provided to citizens and at both administrative and discretionary program what cost. expenditure by organization. Beyond the budget speech, little information is Little coordination with the provinces. The made public. NPC now receives a more detailed overall targets of the budget circular are not binding for provinces. The 1994 budget law budget proposal-this information was restricted ng fo provinc T 1 budget law requires approval of provincial budgets only by to the standing committee-but budget the local People's Congress, in contrast to discussions remain behind closed doors. Most of previous central approval.5 The central level only these discussions take place in January and checks whether local budgets abide by national February, before the official submission to laws Coordination does take place during the Parliament in March. The discussions with the fall finance and planning conference where standing committee sometimes result in changes planning and budget policy is discussed, and in the final budget proposal submitted by the provinces find out what level of central support State Council. they can receive. However, the agreements made are not binding, and only during the fiscal year Separation of Planmand Budget. The continued do provinces find out what actual support is separation of the Plan-managed by SDPC-and forthcoming. Similar to central levels, provinces the budget-managed by MOF-further dilutes are not bound by their approved budgets, as long focus on policy priorities. Although regular as additional spending is financed with additional meetings between MOF and SDPC take place during budget preparation, for line ministries, the two institutions represent two sources of funds ' This topic is the focus of the report: China: Public Investment and that need not necessarily be coordinated. Finance, World Bank report No. 14540-CHA, Washington DC 1995. Moreover, since line ministries do not present an s This also contrasts with the practice of many unitary countries overall budget, their accountability for the results such as China, in which central government has the administrative means to restrict local government spending. However, China operates defacto much like a federal country. 40 revenues. Since local governments have no Budget Law prescribes budget adjustment interest in macroeconomic stability, they procedures for such cases, but they have not been willingly spend the extra revenue, and much of followed. For instance, the rule that additional the in-year overspending is local. expenditures should be financed from savings is never followed. The consequence is that in-year China's murky division of labor among levels of proposals do not compete with other proposals, government creates problems for implementing and approval depends largely on the availability national priorities. China's unitary government of funds. This practice-strongly encouraged by leaves little policy responsibility to local the absence of organizational spending government, but most of spending is local., In limits-further undermines the link between practice, however, policy and administration are policy priorities and funding. Also, the practice hard to separate, and the lack of clarity in comes at the cost of Government's goal to responsibilities leaves policy priorities unfunded. eliminate the budget deficit: if all above-budget revenues for 1996 would have been saved instead Budget year. The budget is not approved until of spent, the budget deficit would have been all well into the fiscal year. The fiscal year coincides but eliminated. with the calendar year, but since budget approval only takes place in March, the budget year is Lack of Fungibility in Extra-Budgetary Funds. effectively reduced to nine months.7 In practice, EBFs are not budgeted for in an annual budget this means that some local spending units do not process, but are instead allocated, as they become have their budgets approved until May of the available. According to the new procedures of current fiscal year. SC circular 29, MOF has to approve the proposed expenditures. But the Comprehensive Such late approval adds to uncertainty of funding Department in MOF, not the Budget Department, at agency and local level, to the detriment of the approves the proposals outside the regular budget ability to plan and deliver services efficiently. process, which is responsible for budgetary allocation. The lack of coordination between the In-year budget adjustment. Even after NPC two departments on the allocation of budgetary approval, budgeting is far from over. MOF has a and EBFs further dilutes priorities. Thus, very conservative approach to revenue although many of the over 900 types of funds are projections, and in every year since 1978 likely to have outlived their purpose, the budget revenues have been more than budgeted, up to 18 process cannot reallocate the resources. (Chapter percent higher in the mid-1980s, but in 1996 still 3). 7 percent more than budgeted. This implies that some 40 percent of resources are allocated Mandatory Spending Increases. The recent outside the regular budget process, assuming that practice of approving laws with mandatory the budget process only allocates additional spending increases outside the budget process revenues. For example, of the Y112 billion undermines fiscal discipline, while not revenue increase in 1996, only Y64 billion was necessarily improving budgetary outcomes.8 budgeted for in the NPC approved budget. The Laws including expenditure provisions, such as the Education and the Agriculture law (Box 4.2) 'Article 107 of the constitution describes the administrative ensure spending increases in priority areas, but responsibility of local governments, but does not delegatepolicy. without detailing the policies the money is used Article 107 states: Local peoples' governments at and above the county level, within the limits of their authority as prescribed by for. In fact, the Laws undermine the incentives law, conduct the administrative work concerning the economy, for agencies to develop effective policies and education, science, culture, public health, physical culture, urban and rural development, finance, civil affairs, public security, control expenditures, because they are nationalities affairs, judicial administration, supervision and guaranteed increased funding. Moreover, family planning in their respective administrative areas. The budget law (Art. 43) allows expenditures not exceeding those Thus contravening Art. 47 of the implementing regulations of the of the previous year until the budget is approved. budget law. 41 Box 4.2: Mandatory Expenditure Increases in Laws and Decisions Function Laws and Decisions Spending Target Education Central Government Decision on Reform of the Education System, May 27, 1985 Compulsory Education Law, approved on April 12, 1986 and put into effect on July 1, 1986 4 percent of GDP by Yr 2000 General Guidelines for Education Reform and Development, February 13, 1993 Teachers Law, approved on October 31, 1993 and put into effect on January 1, 1994 Education Law, approved on March 18, 1995 and put into effect on September 1, 1995 Agriculture Agriculture Law, July 2, 1993 Spending growth higher than Agriculture Technology Spreading Law, July 2, 1993 revenue growth Propaganda and Central Government Decision on Some Important Issues of Expenditure Growth no Culture Strengthening Socialist Spiritual Civilization Construction, lower than overall revenue October 10, 1996 growth Science Science and Technology Development Law, July 2, 1993 Spending 1.5 percent of GDP State Council Decision on Speeding Up Science and Technology by Yr 2000 Progress, May 6, 1995 Environmental State Council Decision, August 3, 1996 Spending 1.5 percent of GDP Protection by Yr 2000 Health Care State Council Decision, January 15, 1997 Spending increase not lower than revenue growth; per capita spending from Y2.6 to Y 4 by Yr 2000 Civil Service Civil Service Law, approved on April 24, 1993 and put into Wages similar to SOE wages Wages effect on October 1, 1993 Recent Budgetary Reforms priorities change over time, but laws are not China has recently undertaken several steps to easily changed, and the mandatory funding. earsiy chnge, an . treform its budget system, focusing on the legal framework. The 1994 Budget Law, further The laws could threaten the Government's detailed in the 1995 implementing regulations, goal to eliminate the budget deficit at the end specifies roles and responsibilities in the budget process, and restricts government borrowing of he nith iv-yar pla, ecus from the Central Bank. The authorities have expenditure cuts in a growing number of areas iroduce a ,l Budge,t"osyte hich becmeles fasbl . Fnly adtr introduced a "Dual Budget" system, which become less feasible. Finally, mandatory seats rcuen ad dveom t expenditure increases apply to all levels of ependtes ancurt rfm aimfoa government, strain resources at local levels, and possibly force local governments to revert separate Social Security budget and a State to illegal fund raising. It is ironic that the Asset budget. practice of mandatory spending increases, . ,, which is probably born out of frustration with Experimt with "Zeased Budeting" the udgt prces, frthe unermies his (ZBB) aim to give increased attention to the ve t process fpolicy contents of budget proposals and to all spending rather than the increase over the previous year (Box 4.3). However, the impact 42 Box 4.3: Zero-Based Budgeting in Hubei Making the Budget an Effective Policy Tool Province If the budget is to become an effective policy To move away from the incremental base number technique, several local governments are tool, then the budget process should both experimenting with ZBB since 1993. ZBB puts more discipline policy choices and translates the emphasis on the planned activities of a spending bureau Government's policy priorities into rather than on last year's budget. expenditure decisions that meet those priorities In principle all expenditure proposals are most The reviewed, not just the increment over last year. cost-effectively following proposed Expenditure norms for office expenses are included in reforms show that this can be done by the finance bureau's budget circular. The spending developing a MTEF, supported by changes in bureaus are required to put considerable detail on its the budget process and institutions that link "special projects" in the budget draft, which allows the policy with spending. These reforms build on finance bureau to probe the proposed expenditures more thoroughly, and to better assess the expected outputs of the experiments in multi-year rolling budgets the bureaus. According to the local authorities, their and ZBB, and should result in an integrated budgeting practice also puts more emphasis on approach to planning, policy, budgeting and controlling personnel costs, as the more detailed budget evaluation, and the systems and procedures to proposals allow for a check of actual personnel rather support it. Figure 4.1 illustrates the planning than the quota of personnel assigned by the Personnel Bureau. and budget process (see Annex 4 for details). The local authorities state that the ZBB technique has enabled them to better control overall The proposed reforms are far reaching, and expenditures, control wage costs, and readjust can only be implemented over time. China's expenditures in priority areas. However, they feel that the technique has its limits, because a large part of the approach to reform-gradual and based on budgets consists of personnel costs ("meal finance") experimentation before nationwide which cannot be altered through the budget process. implementation-is well suited for the of the technique is limited, because the format countries' decentralized budget system. of the budget proposals in the experimental However, OECD experience shows that localities suffer from the same lack of detail as budgetary reforms require strong and ongoing normal proposals, and fail to specify expected political commitment, and a change in culture outputs and outcomes. Moreover, the within the public sector and with Government's accounting system and the policymakers. Reforms only work if the budgetary classification system provide policy resulting budget process disciplines makers with only limited information on the policymakers in their decisions. costs of budgetary proposals. Finally, personnel costs, which make up a large part of Ensuring Macroeconomic Consistency the budget, fall outside the budgetary process. Many policy decisions only have their full budgetary impact over time. To ensure The MOF comprehensive department has been spending consistent with the government's experimenting with a five-year rolling budget deficit targets and macroeconomic stability, plan. The plan is not operational in the budget China's budget system needs to move to a process, and is developed independently from multi-year framework. Such a framework next year's budget plan, in part because the requires: information for the multi-year plan differs from that of the annual budget. No shared * Multi-year revenue projections based on databases in MOF exist, and outside data for macroeconomic conditions, and existing developing the plan are also transferred and future tax policies; manually. Multi-year projections and annual * Multi-year projections of sustainable or projections alike are hindered by the absence politically acceptable deficits; of an economic classification of expenditures. 0 Multi-year expenditure requirements for funding existing policies. 43 Figure 4.1: Linking Planning and Budgeting (1) REVIEW the previous planning and implementation period (6) (2) EVALUATE and AUDIT UNDERTAKE PLANNING activities' effectiveness and feed ACTIVITY the results into future plans Establish resource framework, set out objectives, policies, strategies and expenditure priorities (5) (3) MONITOR activities and MOBILIZE AND ALLOCATE ACCOUNT for expenditure RESOURCES Prepare Budget (4) IMPLEMENT PLANNED ACTIVITIES Collect revenues, release funds, deploy personnel, undertake activities sectoral limits would have to be broken In OECD countries, the above projections are down by ministry and agency for further usually captured in a document termedfiscal detailed budget preparation, and should be outlook or policy framework paper. This included in the budget circular. document, whether internal or public, serves to set the stage for budget preparation, and MOF's experiments with multi-year budgeting identifies the budgetary room available for could become the basis for a multi-year new policies, or the need to cut on existing expenditure framework that would ensure policies in order to meet the deficit targets. consistency of policy choice and Macroeconomic consistency requires further: macroeconomic stability. As a first step, the information base for the projections must be * Expenditure limits that are set at the start improved. In part this could be achieved by of the budget preparation rather than at the integrating the data used for the annual end. The overall expenditure limits would projection with those for the multi-year be linked with macroeconomic conditions, budget. Currently the Comprehensive revenue projections, the costs of existing Department, which is in charge of the multi- policies and an acceptable deficit. year projections, does not share data with the Budget Department that compiles the annual * Sectoral expenditure limits set at the start budget. of the budget process based on policy priorities of the State Council. These 44 Second, sound multi-year projections require the MYEF requires consensus on the policies detailed knowledge on the composition of that will be adopted within those limits. This expenditures. The current budget classification requires revamping of the budget process. and reporting system therefore need to be reformed. The system should allow for a Linking Strategic Policy Choice and the breakdown by economic category, to enable Budget analysis of the impact macroeconomic China currently lacks a mechanism that conditions has on ongoing expenditures. For translates the broad priorities of the State instance, the budgetary consequences of an Council into spending on policies that work. increase in wages, or rising prices for building Such a link requires creating a step in the materials should be easily traced. MOF should budget process that actually makes that link, also be able to distinguish between statutory more detailed budget proposals to establish expenditures that are hard to change without that link, and more scrutiny of budget changing laws, and discretionary expenditures that could be reallocated within the budget process. From the perspective of strategic priorities and operational performance, it is Institutionalize the link of policy and budget. essential to also have a China could improve the choice among functional/programmatic classification to focus competing policy proposals by better attention on policy priorities, their cost and integrating this choice within the budget implementation. A more sophisticated process. Rather than leaving policy choices to classification system could make such a low level officials during budget negotiations, distinction, and fully clarify choices for policy- China could consider making these choices in makers. a high-level ministerial committee. One means of doing so is an expenditure review Third, MOF needs to improve revenue committee. Such a ministerial-level committee projections, and extend them beyond the one- could, on the basis of the overall expenditure year horizon currently used. Most of the ceiling and the expenditures for ongoing revenues come from taxes, for which SAT still policies, decide which new policies best meet makes projections. However, the MOF tax the State Council's strategic objectives. policy department is currently building the An expenditure review committee, or a similar capacity that will improve revenue projections. body that links priorities with spending, could To effectively integrate these projections in the be in charge of breaking down the aggregate multi-year expenditure framework (MYEF), spending limits to ministerial and agency these projections should be widely shared spending limits. Budget negotiations would within MOF. then be restricted to checking whether the various ministries have indeed included the Expanding the experimental multi-year budget policy decisions of the expenditure review to a full-blown MYEF will require committee into their budget proposals. Since considerable capacity building in MOF and the ministries would be required to remain line ministries. Even if MOF is in charge, a within their assigned spending limit, the MYEF can only become a credible vehicle for endless negotiations on individual spending budgeting, if line ministries fully understand proposals could be abolished. Expenditure the process and agree with the estimates. As a review committees (or inner cabinet or cabinet start, MOF could include multi-year estimates subcommittees) have been powerful tools in the budget documents, without aligning the budget with policy in countries operationalizing them as a decision tool. Over such as the UK and Australia time, the estimates should become the starting point for the budget discussions. Enforcement of the overall spending limits resulting from 45 The proposed expenditure review committee achieve, and their link with Government's could have several sub-committees along priorities; sectoral lines, to overcome the disadvantage of * the expected outputs of the policy, how the compartimentalization and narrow scope of they link with the outcomes, and how these the Chinese ministries. For instance, sectoral outputs can be measured and evaluated; review committees for infrastructure, * the specific instrument appropriate for the agriculture, and education, which would policy (produce the service; contract out to consist of ministers from those sectors, plus the private sector; subsidize an non- the Finance Minister. government organization (NGO) to deliver; fund the beneficiary directly); As a second-best alternative to an expenditure * multi-year estimates of cost and review committee, a stronger link of policy expenditures for the policy, the required and budget requires involvement of MOF in budgetary contribution, and the other every policy proposal that will have budgetary sources of finance. consequences. This involvement should be, irrespective of whether the proposal is made Build Analysis and Evaluation Capacity. To within the context of the budget process, or ensure that the policy proposals that are tabled whether the proposal is made during the year. in the budget process are sound, China would Having the Finance Minister as a permanent need to strengthen the government's capacity member of the State Council would ensure that to analyze and evaluate budget proposals. fiscal considerations are taken into account in Analysis of budget proposals should ensure policy decisions. However, a disadvantage of that the proposed expenditures meet the this approach compared to an expenditure Government's priorities, by funding those review committee is that is would perpetuate activities that give the outcomes desired by the the adversarial relation between MOF and all State Council. Expenditure evaluation does other ministries, because the Finance Minister the same for ongoing spending, and it can be a would be the only one to ensure that aggregate powerful tool for identifying savings options fiscal targets are met. that would facilitate reallocation of budgetary resources to high priority sectors. Some Approve line ministry budgets. A necessary capacity for analysis and evaluation already step for financial discipline is to break down exists in the ministries, but the results of these the sectoral budget proposals into line ministry efforts do not yet play a large role in the budgets. As a first step, the organizational budget process. Much of the analysis and budgets could be presented as an annex to the evaluation of policy would have to be done by budget speech. But accountability would be the ministries themselves, but MOF can play a greatly increased if each ministry would leading role in setting standards, defining present its own ministerial budget to the NPC. procedures, providing training, and controlling Make budget proposals more detailed To the quality of the analysis. MOF could also achieve better links between strategic priorities jump start analysis and evaluation, by taking and the budget, funding proposals need the lead in thorough analysis of budget considerably more detail than at present. proposals and evaluation of ongoing Ideally, each spending proposal would include: expenditures. * a justification for Government Ex-post evaluation can be supported by value intervention, why the funding proposal is for money audits of the China National Audit the best instrument of intervention, and Office (NAO). The Audit Authorities still need why alternative intervention mechanisms to further build their capacity for financial (regulation, devolution to other levels of government) are not chosen (Box 4.4); * the specific outcomes the policy tries to 46 Box 4.4: The Changing Role of Government Considerations on the role of government should become an integral part of the budget process and policy selection. At present, China implements most of its policies by direct provision through a department, an NPO, or an SOE, or regulate on a case-by-case basis, for example through investment approvals. This leaves many options hardly used. In the move to market economy, China is likely to increasingly leave activities to the market by not intervening. And regulation of the market as a policy instrument will increasingly replace direct intervention. When markets mature, contracting out of services-either to SOEs or to private suppliers-becomes more and more an option. Finally, China makes as of yet little use of funding beneficiaries directly. The OECD reforms over the last decades show that widening the choice of policy options-including non-intervention-can greatly increase government's performance. The budget process can be a powerful tool in reshaping the role of government: government is what government funds. By requiring that budget proposals include a motivation why government should intervene, and why it should do so by spending rather than by other means, the role of China's government can adapt in a gradual fashion to one more compatible with a market economy. Government Intervention Provide Govemment Directly Enterpasnes Dont Intervene < 3 Departmentment Policy CoicReta Fnd Consumer Interene .- Regulate 33Vouchered Specific Purp Entity =DevoWe uothority_Y Sobrdinte Feliny Cheice i6T's,Pff r u t4f by 1980s, several OECD countries have made a budgetary agencies would be expanded special effort in reviewing their existing (Chapter 5). But over time, the NAO would policies (Chapter 3). have to increasingly focus on value-for-money audits, rather than only on the compliance Improve Policy Coordination audits as is done currently. Coordinate closer with SDPC. A better link Over time, evaluation and value-for-money between policy, planning, and the budget audits would have to become an integral part requires increased coordination among MOF audts e normalbudgt beceanineg art oand SDPC beyond the macroeconomic of the normal budget process. One way of dialogue now taking place. The integration of doing so is by including evaluation plans in EBFs in the budget makes such coordination each budget proposal. But in the short run, even more urgent, as these funds now cover China could choose to accelerate moving away most of the recurrent costs for investments from the inherited spending patterns. In the from the plan. As a first step, the multi-year 47 budget of MOF should be fully linked with the and finance ministry, but combined these rolling investment plan of SDPC. This link responsibilities in the MOF. would allow for budgeting for recurrent costs associated with investment projects. Such a Coordinate within portfolios. China's policy link would also allow for better multi-year projctins, nd hus or ore ccuate priorities could be better served by budget projections, and thus for more accurate pooasta oe h opeepirt determination of the resources available for sector. th er th having te irio .e plc. sector. Rather than having the ministries of agriculture, water resources, and forestry each d .e submit its own proposal to MOF, their Soen since SDPCul till playrnmes t oatc proposal could be merged into one covering all role in formulating government's strategic oagiutr. Siia. osdrtoshl o prioitis, DPC houd pay leaingrol in of agriculture. Similar considerations hold for priorities, SDPC should play a leading role in tetasotsco,wihi avdu the transport sector, which is carved up the expenditure review committee(s), once they are established. However, its role in construction, ind ommuniao detailed ministerial-level planning should comttees of t oe xniture diinsh o btinoneshp.n Subcommittees of the proposed expenditure dimiish To obtin onerhip and review committee could be instrumental in accountability for the line ministry, sectoral rmulating cud prosal fr planning should become the task for line ministries themselves. They should be priority sector. responsible for formulating the policy Over time, and in line with the ongoing civil initiatives necessary to meet the Government's priorities within the budget limits set by the eve reorisTis ould bm or even abolished. This would not only bring expenditure review committee or a similar China's goverment administration more in body. line with the emerging role of the state, but also in line with its policy priorities. Merging Thsponsbiiird SDPC fic all shet o would facilitate better trade-off of competing rnesonbilites fin magovenment of policy choices within the priority sector. For invetmen prject wihin ovenmen to instance, the ministries for agriculture, MOF, and relinquish completely control over .ntne h iitisfrarclue imOF,atreinqi ompley ontover t forestry, and water resources-each involved in investments. agricultural policy-could be merged to emphasize the priority government puts on that Fourth, when over time the size of the plan has sector, and to better allocate the increasing diminished, the plan could gradually be resources for that sector. The announced integrated fully into the budget presentation. reforms already make a big stride in abolishing SDPC could still present the overall strategic those ministries directly involved in productive goals of Government to NPC, but the budget activities (Box 4.5). would include all financing and spending for Government, whether recurrent or capital Coordinate better with provinces. China's spending. Line ministries could then present highly decentralized fiscal system requires thendeig.Lne udgetriter corganiatin. effective macroeconomic coordination with the detailed budgets for their organization. provinces. Such coordination would gain in Finally, in the long run, Government could importance once provinces will be allowed to consider merging MOF and SDPC. Korea, borrow from the capital market-currently which had for 40 years a separate Finance and prohibited in the budget law. But even without Planning Ministry, took this step in 1994. borrowing, the provinces' procyclical spending Japan, which has had a considerable dose of has macroeconomic consequences. China planning through the fiscal investment and therefore needs to consider coordination loan program never had a separate planning mechanisms that are more effective than the current planning and finance conference. Of 48 interest are the models of Germany and Australia, which coordinate with subnational Box 4.5: OUTLINE OF CHINA'S governments in a multi-year framework (Box GOVERNMENT DOWNSIZING 4.6). As a minimum, effective coordination requires: China's government downsizing approved by the National People's Congress: * working on a common set of assumptions MINISTRY-LEVEL BODIES TO BE ELIMINATED On key macroeconomic variables, Ministry of Power Industry Ministry of Coal Industry including growth rates, prices, inflation, Ministry of Metallurgical Industry relative price changes, wages and Ministry of Machine Building Industry Ministry of Electronics Industry Ministry of Chemical Industry * working with common revenue and Ministry of Internal Trade expenditure classifications; Ministry of Posts and Telecommunications ec Ministry of Labour * producing budget forecasts early on, and Ministry of Radio Film and Television well in advance of the budget year, so that Ministry of Geology and Mineral Resources Ministry of Forestry thorough discussion can take place, and State Physical Culture and Sports Commission the benefits of coordination can be State Commission for Science, Technology and integrated in the final budget proposals. Industo State Commission for Restructuring Economy Beyond the macroeconomic framework, NEWMINISTRY-LEVEL BODIES cosState Commission of Science, Technology and increasing policy coordination at the strategic Industry for National Defence level could more effectively link the State Ministry of Information Industry Council's priorities and the budget. The Ministry of Labour and Social Security Ministry of Land Resources Finance and Planning conference comes too late in the year to have a meaningful impact on MINISTRIES WITH CHANGED NAMES budgetary decisions, and additionally policy State Planning Commission will be renamed State Development and Planning Commission coordination among levels of government is State Science and Technology Commission will be called for (see also Chapter 5). renamed Ministry of Science and Technology State Education Commission will be renamed Ministry of Education Move Away from Incrementalism MINISTRIES, COMMISSIONS AND OTHERS Improving the link of strategic policies and the THATREMAIN UNCHANGED budget process will take considerable time. Ministry of Foreign Affairs Ministry of National Defence However, much scope exists for China to State Economic and Trade Commission further move away from incremental State Nationalities Affairs Commission budgeting in the short run. Measures that can Ministry of Public Security budgtingMinistry of State Security be taken include: Ministry of Supervision Ministry of Civil Affairs Ministry of Justice Ministry of Finance Ministry of Personnel Ministry of Construction Ministry of Railways Ministry of Communications Ministry of Water Resources Ministry of Agriculture Ministry of Foreign Trade and Economic Cooperation Ministry of Culture Ministry of Public Health State Family Planning Commission People's Bank of China Auditing Administration Source. Reuters 49 Box 4.6: Intergovernmental Macroeconomic Coordination in Germany and Australia In Germany, the budget is planned within the framework of a rolling five-year financial plan. These plans are prepared both by the federal government and the Laender (states). The Financial Planning Council then ensures that these plans are coordinated and that they are consistent with the aggregate fiscal target. The Council is chaired by the Federal Minister of Finance, with finance ministers from each of the Laender and representatives of municipalities as members. The Committee on Public Borrowing, which has a similar composition, and coordinates public borrowing and ensures that capital market demands are not destabilizing, complements the work of the Council. The Council decides on the overall level of borrowing, whereas the Committee decides on phasing of borrowing. The results of the deliberations are a Federal Financial Plan comprising of a set of compatible financial plans for lower levels of government. Based on this financial plan, budgets are prepared. Supplementary allocations are rare at all levels. In Australia, before the Premier's Council decides on the overall level of transfers to the States, the Treasury prepares a memorandum for the Cabinet that outlines fiscal developments and prospects in the state and local sector. These are prepared in the context of the forward estimates of outlays for general- purpose assistance and the Commonwealth's broader objectives for public sector spending and borrowing. Source: Barry Potter: Budgetary and Financial Management, in: Teresa Ter-Minassian, Fiscal Federalism in Theory and Practice, IMF, 1997. * Budget proposals could be made to Within existing overall quotas, this would, compete with existing policies by among others, require simpler procedures requiring agencies that propose additional for changing personnel files among depart- expenditures to propose savings options as ments, and more retraining within the well. In the UK this is common practice, administration. and Australia has a similar system, in * Mandatory spending increases in laws which the Cabinet can choose to accept the should be avoided. Such increases may new policy proposals, or to only accept the thwart efforts to bring expenditures under savings proposals of the ministry; control, and reduce the deficit. Once a * Government could consider introducing MYEF is in place, such proposals can be explicit time limits on funding decisions. scrutinized like any spending proposal, but This so-called "sunset legislation" has until then, MOF should have a say in any been used in Japan and other countries, of such proposals made. Beyond explicit and has the advantage that policies have to consideration in the budget process, MOF be renewed by an explicit policy decision, would have to carefully monitor at which time the effects of it can be mandatory spending and entitlements in evaluated; the budget. If the share of such * Agencies could be required to deliver an expenditures becomes too large, the "efficiency dividend" for certain budget loses much of its flexibility, categories of expenditures, for instance thereby severely restricting its use as a running costs. Over time, with a change in macroeconomic policy tool. personnel allocation policies, personnel * Minimize in-year adjustment. To maintain costs should be included in such a system. the integrity of the policy choices made at (see Annex 8: The Running cost system in budget time, in-year adjustments should Australia); become exceptions rather than the rule. * Even without reductions in overall Better revenue projections would help in numbers, governments at various levels allocating a larger proportion of resources could focus on increasing mobility of at budget time. personnel among functions and agencies, * Such projections would also take away the thereby avoiding the costly redundancies incentive for ministries to continue to that perpetuate the incremental budgeting. submit budget proposals during the budget 50 year: once the annual budget allocates all * Parliamentary budget deliberations, so that resources, there is no use in applying for the budget could be approved before the more. In-year budget adjustments could start of the budget year. China should still take place in exceptional consider whether it wants to move the circumstances, for instance when revenues budget year, or whether it wants to move are unexpectedly high, or when NPC deliberations. Since the NPC plenary expenditures need to be cut to avoid session has traditionally been held in inflation. In such cases, in-year spring, it seems unlikely that these will be adjustments should follow the rules of the moved. This leaves moving the budget Budget Law. The Budget Law prescribes year to, say, April 1-March 31. The that in-year adjustments should broadly disadvantage of moving the budget year is follow the same process as the budget that it no longer coincides with the itself, except for NPC approval. If this calendar year, which then requires separate process is followed, competition among macroeconomic projections for the budget proposals is assured, and dilution of policy year. However, many countries, including priorities is avoided. Japan, the United Kingdom, the United * Realigning the budget year. Finally, the States and Australia have budget years that budget year could be brought in line with differ from the calendar year (Table 4.2). the budget preparation calendar to increase predictability of funding for spending agencies and localities. Australia, after approving the budget well into the budget year for decades, four years ago moved the 51 Table 4.2: Current and Reformed Budget Process Current budget process Tentative new budget process after reforms * April Discussion of strategic priorities in State Council. (year-1) Results disseminated to ministries, agencies and provinces * May Discussion of strategic priorities and tentative budgetary implications with provinces * June Central government and local governments update multi-year estimates. Fiscal outlook document produced by MOF State Council determines overall spending limit for next three years * July Evaluation reports of ministries tabled. Main savings options identified * August Expenditure Review Committees meets, sets sectoral and ministerial expenditure limits for next three years. * September Sectoral expenditure review committees meet, determine main savings, main new policy initiatives * November Mid-year review of the budget. Multi-year expenditure estimates updated Fiscal Outlook Document updated Expenditure Review Committee determines key policy and spending changes. * October- State Council orders drafting of the budget * November Key policy and spending changes discussed with November and sets out broad guidance for the draft provinces. Prospective grants identified; draft provincial budgets discussed * October- Spending Units, Ministries, and local December governments compile their draft budgets, and submit to the sectoral departments in MOF (In localities, some proposals do not come in until February of the budget year). * November MOF consults with Ministries and local governments on next year's budget * November National Finance Conference; MOF drafts - central budget, and compiles State budget December * January Fiscal Year Starts * January Multi-year estimates for line ministries and sectors finalized State Council decides on next year's budget proposaL Submits proposal to Standing Committee of the NPC * January MOF submits draft Central Budget and State budget to the State Council. Discussion in the State Council on outstanding issues and major projects. * January- Audit of Spending units' accounts by the * February Deliberations of the Standing Committee March NAO. * February Presentation ' of Draft Budget to the Standing Committee of Finance and Fconomics of the NPC * March Presentation of the Central and State * March Main policy changes, budget, and draft final accounts Budget and the preliminary Final presented to NPC, and published. Sectoral and Line Accounts by the Finance Minister to the Ministry budgets approved. NPC * March Provincial Budgets approved * March- MOF passes on the approved budgets to * April Budget Year starts May the spending ministries. Spending ministries pass on the approved budgets to the spending units * January- NAO audits the central budget, reports to * April-June Audit Office audits financial statements of budget April State Council units and central budget. Reports to NPC, publishes reports * June-July Presentation of Final Accounts by the Finance Minister to the Standing Committee of the NPC 52 CHAPTER 5 IMPROVING INCENTIVES FOR PUBLIC SECTOR PERFORMANCE China's decentralized government structure could be a major strength in achieving public sector performance. Higher performance requires clear assignment of functions, more autonomy in policy implementation and resource use, a shift from compliance control to accountability for results, and information systems that monitors performance and enforce budget constraints China's Decentralized Government Government departments, and are subject to the government's budget procedures and personnel China's government is highly decentralized by regulations, leaving little initiative to the region and function. It has 5 levels of government (Figure 5.1), with a total of some Figure 5.1: China: Levels of Government 50,000 entities. A growing number of government employees work at the local level ce (Figure 5.2).1 Subnational governments spend Provinces M.nicipatitiesiT11 over 70 percent of budgetary resources, of which 27 4 about.half is financed by central grants. 7e---------- Prefectures Cities under Provinces 127 206 Functionally, China's central government has 40 ministries and commissions directly under the cioeesunderPrefectures State Council, supplemented by 43 bureaus, institutions and administrations. Below these Townships Towns entities, about two thousand NPOs deliver most of central government services. Local line aistis Urban Districts 2 I F 697 bureaus mirror much of the central structure, and an astounding total of over I million NPOs Village Committees Neighborhood Committees delivers services and employs much of the 802,100 government workers. Education accounts for 1/ Beijing, Tianjing, Shanghai, Chngqing almost half of employment, followed by health 2/ Under Cities at all levels. care and natural resource management. NPOs Source: State Statistical Yearbook are accountable to line ministries/bureaus and rely on a mix of budgetary support and EBFs.2 mnager e laconcleareols an inadequate reporting and accounting results in weak accountability, for results, nor are the The decentralized nature of China's government rwrsfrcvlsrat ikdt civn could be a major strength. Service delivery close tes results.sAnd nsicie copetitin to the final user can better take their preferences the nstte sector ee m pe for the non-state sector leaves much scope for into account, exploit local differences in costs and tastes, and stimulate innovations that further the Government's policy objectives. However, Functions are Unclear China's public sector managers do not have the autonomy, or the accountability, to exploit this stuturey Mn N s ac o raite mu eplike China's incomplete transition from a planned to a structure. Many NPOs operate much like makteooyhslcetda mxo market economy has created a mix of commercial and government functions at agency Except for 1994, when the national tax administration was levels. Because the division of responsibilities established. 2 According to the ministry of personnel, 60 percent of NPO personnel are fully financed from the budget, 23 percent partially, and 17 percent relies fully on EBFs. 53 Figure 5.2 :The Bulk of Public Sector Personnel is in NPOs at the Local Level Public Sector Employment, 1994 Government Cadres by Level of Government, 1994 Military 3 millions Central 2% Government: NPOs: Government O i on 10.3 mlions 24.5 millions Townships 9% Provinces 21% 7 17% 0 oPrefectures Coerities 25% 43% SOEs 75.5 milions 52% Government Cadres 1982-94 Distribution of Personnel in NPOs, 1990 5% 2 34millio 3% 6.000 203 IIlon 0 1982 194 1 - 98 19 91 I 2 19 19a Note: Government Cadres are the equivalent of civil servants Source: Ministry of Personnel inheritance from the Maoist period-there is These commercial ventures may generate overlap and waste. conflicts of interest, but more importantly, they divert management attention and resources from Mix of Government and Non-Government. the main purpose of the organization. This leads China's NPOs are a very mixed bunch. Some, to inefficient delivery of the public goods the such as schools, hospitals, and research institutes agencies are meant to produce. They also give have a clear role in the delivery of public goods, government agencies the incentive to divert and are mainly funded by the budget. Others, public funds in commercial ventures, as these such as seed management bureaus, veterinarian's generate the revenues that pays some of the services, and grain bureaus run commercial salaries and most of the perks for the civil operations. The drive for commercialization servants. since the mid-1980s has increased cost recovery for many government services-witness the Concerning government functions, NPOs show exploding EBFs. At the same time, it led NPOs, little understanding how their activities fit in with and even ministries to enter in purely commercial Government priorities. Their budgets are of services. School-run enterprises are little use as a guide, because these predominantly common-and even stimulated by tax credits; the contain financial data, not goals and purpose of State Administration for Taxation runs expenditures. Business planning and strategy commercial tax advisory services and a hotel; formulation does not seem to take place in the and the MOF runs a Trust and Investment NPOs, but is left to the supervising ministry, Corporation. 'Source: Mission interviews 54 which seems to have a pervasive influence in the Autonomy of Managers day-to-day operations of the NPO4 Expenditure discretion limited, and misused. The Unclear expenditure assignments. Unclear budget law (Art. 56) and its implementing Unclarexpenditure assignments . Unmaor regulations (Art. 62) allow spending units to expenditure assignments remain a major transfer among budget items only after approval weakness in China's intergovernmental fiscal by MOF (or finance bureaus at local level). The relations. On paper, China's intergovernmental regulations do not specify an amount under division of responsibilities is roughly m line with which the budget unit itself could reallocate, nor that dictated by public finance theory and does it specify the conditions under which such reflected in international best practices (seedosispcfthcniinsuerwchuh refleed in). Interatonl es practice s (ese o approval is given. In practice, budget units often Annex 10). In practice, several levels of do not seek approval for virement, and the government are involved in the same or similar Finance Minister's repeated remarks on funds activity. Much of the overlap results from the diversion and abuse (see Chapter 1) indicate that Maoist policy of self-sufficiency of each locality, reallocation is often not in line with the Each province and even each municipality was Government's goals. supposed to be more or less self-sufficient. Carryover does not pay. Art. 23 of the Unfunded mandates. Central government implementation regulations of the budget law regularly decides on policies that affect spending at the lower levels, but do not provide the alwfrcr~oe f"pca ups"ies fiancial t eansor thvels o Renot pre te Apparently, operating costs are excluded from f. this carryover rule. The article prescribes that a of these unfunded mandates include: the nine- year compulsory education; the grain pricing and sru sud b l the bdtry and he dcisin t serice age reserve fund and the necessary expenditures in procurement policy, the 1993 civil sthe following year. Thus, savings budget units reform; acreate pension have little incentives to save on their pooling for SOE workers at the municipal level, expenditures, as they will lose their budget The grain policy alone cost the government of allocation to the budgetary reserves, which are Jilin Province more than Y2 billion in trading still well below the aspired level of 5 percent of losses in 1996, or one-quarter of the provincial budgetary revenues. The policy, which expenditures. guarantees farmers a procurement price, caused Personnel management is inflexible. Line the losses when a bumper harvest caused market ministries and agencies have limited autonomy prisciesndagncesoae fmtelaloom prices to fall. over personnel allocation. The State Committee at for Public Service Structure under the State Localrgoenmnth s cliner tatis "t cenl Council, assisted by the Ministry of Personnel government throws dinner parties but leaves local governments to pay the bills." Unfunded determines how many people a ministry or an mandates explain much of the disparities in agency can employ. It also sets overall limits on local levels of government, which are further service standards among regions: whereas richer regions can easily finance good health care and spcfe by lca au i he pese eductio frm thir wn eans porerquota normally acts as an upper limit, because eductio frm thir wn eans porerpolitical pressures force government agencies provinces do no have sufficient funding to meet polit preres emgovernment a es and units to serve as employer of last resort. the central standards, nor do they obtain much in Line managers also have y Linemanaers lso avelittle incentives to save earmarked grants to do so. on personnel, as this is separately budgeted for. If they would save positions, they would simply lose budgetary allocation. This impression is based on mission interviews with NPOs and their supervising ministries. 55 Expenditure Monitoring and Control annually to MOF, both on its own operation and Expenditure monitoring is perhaps the weakest on the operations of treasuries at provincial levels. Daily and monthly reports consist of link in China's budget process. Much effort is spent on ex-ante control, but actual expenditures aggregate numbers on revenues transferred to the are not monitored by the treasury system. treasury account, transfers made from the Insufficient monitoring risks abuse of funds, and account, and the treasury account balance. yof managers. Accounting system and budget classification. China's accounting system is cash based, and as such cannot record cost. Assets are recorded, but Treasury does not monitor expenditures. China's ti sdn eaaefo h ugtr treasury management system, adapted from the accunin se.rte sedn ts record accounting system. The spending units record Soviets in the 1950s, does not monitor transactions in a great amount of detail, but most expenditures directly. The system records theg traesferuofsfundscfrom the cystena treaordsur of this is lost in aggregation at the ministry level. transfer of funds from the central treasury Sedn nt eott hi uevsn accont o mnisty lvelbankaccunt in Spending units report to their supervising account to ministry level bank accounts in miisr . rbra,wihcnoiae h commercial banks. From there, the money is mnsr rbra,wihcnoiae h commrcil bnks.Fro thre, he one isaccounts of all subordinate-spending units. The transferred to bank accounts of the spending unit, account ot s nateendin unitthe from which spending takes place. Reconciliation agreate repor.ten with bank account statements only takes place with the final accounts, after the budget year has The accounting classification is a mix of closed. functional and economic classification. Some In contrast with expenditures, the spending unit's items-such as subsidies to SOEs-are recorded as requests for funds that spending units submit a subtraction from revenues rather than undergoes elaborate scrutiny: at spending expenditures. Interest payments are recorded ministry level; at the sectoral department in below the line, thereby causing an underestimate MOF; at the accounting division of the budget of the budget deficit. The cash basis of the department in MOF; and at the treasury accounting system precludes adequate recording department of PBC. However, the scrutiny rarely of arrears. Arrears are regularly reported in grain ever leads to changes in the request for funds, procurement and teacher's pay, but perhaps the and the decisive factor for release of funds from largest arrears accrue in the value-added tax MOF is whether sufficient funds are available. refund. At the end of 1995, the amount of arrears was reported to be over Y80 billion. This implies The treasury management system results in a that the budget deficit on an accrual basis was large float of government money in the banking almost 1.5 percentage point of GDP larger than system. Moreover it leads to much delays in the cash-based deficit of 1.7 percent.5 disbursement of funds to spending agencies, and possibly to payment arrears. At the same time, Final accounts have financial information only. the system provides little effective expenditure China's final accounts provide only limited control and monitoring by MOF, and therefore insights in the State's finances. They include little accountability for line ministries and detailed statements on revenues and agencies. expenditures, but do not report on assets, liabilities, contingent liabilities, guarantees, flow No reporting on results. China has an elaborate of funds, and other reports that are usually reporting system for the treasury and spending included in OECD countries' fiscal reports. No units. Spending units report monthly and systematic reporting on outputs or outcomes of annually to supervising ministries, which compile expenditures takes place. reports for MOF. MOF, however, only substantially evaluates the annual reports. The , See: Garamfalvi et al China: Government Budgeting and Treasury central treasury reports daily, monthly, and Management: Issues and Proposals for a detailed review of China's budget classification. 56 Audit of the emerging private sector. Civil service pay External audit focuses on compliance. Audit in is mainly based on longevity and grade, and China is mainly limited to compliance audit. The shows little relation with performance. Although audit law requires the Audit Authorities to there are bonuses, these are mostly seen as part . of the normal wage, not as rewards for good inspect whether money is well spent, but the erformal limited performance reporting, and the lack of performance. relevant skills in the Audit Authorities prevents . c wiepedvau-o-onyadt..oee,i The civil service wage structure is becoming widespread value-for-money audits. However, the highest-ranking 1996, for the first time, the Audit Authorities moreicomprned In 7 te moresthanking reported directly to the NPC on the central final official earned about 7 times more than the reprte diecty t th NC o th cetra fiallowest ranking, down from over 10 in 1955. By accounts of 1995.6 Before, reports were made to MOF and the State Council. Among the 1993, this was less than five times (Table 5.1). auditors' findings was that budgetary money was Wage compression was necessary to keep the overall wage bill low in light of an inflation in used to speculate on the stock market-a directrak In18,2pectofllivlsvns consquece o wek trasuy maageent.The ranks. In 1982, 82 percent of all civil servants consequence of weak treasury management. The were normal staff; in 1994, this was only 60 Audit Authorities are expanding their capacity w for value-for-money audits, but are still severely Table 5.1: China: Civil Service Wage limited by the information included in budget Structure 1985 and 1993 requests and fiscal reports. (mid-point salary, Yuan per month) Ranks 1985 1993 percentage Internal audit is extensive. Internal audit in change China is extensive. The supervision department Premier 453 1115 146 of MOF checks line ministry and agency Vice Premier 343 984 187 Minister 282 881 212 compliance with financial and tax regulations. Vice Minister 236 769 226 The Ministry of Supervision investigates Director General 189 668 253 administrative misconduct that may result in Deputy Director 165 581 252 criminal cases, and the annual Fiscal and Tax Division Chief 146 489 236 discipline examination campaign does both. The Deputy Division Chief 127 422 232 Section Chief 110 372 240 campaign is a phenomenon specific to China: Deputy Section Chief 94 330 252 during several weeks each year personnel are Section Member 80 302 276 switched from their own unit to another unit to Clerk 67 264 294 conduct a review of the books. The focus of Source: Ministry of Personnel each of these internal audits is again compliance percent. The decline is fully explained by the with regulations, not on performance or value for rapid increase in the number of section chiefs. money. Moreover, there seems to be considerable overlap in the various internal audits, and some A developing civil service system. China is SOEs and State Banks that are also subject to the rapidly developing a civil servant system. The audits complain about the time involved in them. 1993 Provisional Regulations on Civil Service lay the basis for such a system. The regulations Incentives for civil servants aim to: establish competitive examinations at Civil service pay not linked to performance. entry; allow dismissal of incompetent workers; MOF and the ministry of Personnel centrally and to systematize training and career determine pay. Despite several rounds of wage perspective for civil servants. A job rotation reforms, civil service pay lags behind that of system for director level civil servants has been SOE employees, and much more so behind that introduced, and a civil service training school has been founded. The civil service system has been introduced down to the municipal level. * See: Renmin Ribao (People's Daily), July 4, 1996. 57 Competition China makes little use of the nonstate sector for Table 5.2: A Perception of Corruption public goods provisioning. NPOs and SOES Rank Country Score 1996 (max. 100) is Denmark 9.94 deliver the bulk of services, and little is 2 Finland 9.48 contracted out. Although some private financing ... ... ... of energy and transport investment has taken 40 Philippines 3.05 place in prosperous coastal regions, this is still 41 China 2.88 pae42 Argentina 2.81 met with much skepticism of central officials. 52 Nigeria 1.76 Recently, China concluded its first Build- Source: Transparency International Operate-Transfer scheme for energy with private Recent Reforms foreign capital as the main financiers. However, municipal services such as water supply, garbage China has major ongoing initiatives to improve collection, and street cleaning; and social incentives for the public sector. Since 1990, the services such as health and education are firmly administrative reforms have encouraged in the hands of SOEs and NPOs. separation of commercial activities from government through corporatization. Some Procurement rules are weak. China's ministries-such as those for Light Industry and underdeveloped government procurement system for Textiles-have become industry associations, is an impediment against more non-state although they continue to perform some involvement in public goods provision, government regulatory functions. The number of Government procurement often involves simply civil servants at all government levels fell by shopping for the item needed. Competitive bids some 10 percent since 1994. The next wave of are usually limited to SOEs. China's informal reforms targets a reduction of 10 percent in NPO arrangements give rise to waste in budget personnel. Following the Civil Service implementation, and could be a major cause for Regulations, China has introduced competitive corruption. Despite recent improvements, China entry exams, and a pilot wage comparison still ranks low in Transparency International's between civil servants and SOE workers was corruption listing (Table 5.2). Government done, in preparation of further wage reforms. procurement rules have become a bone of MOF and PBC have also started experimenting contention in China's WTO negotiations. In with better treasury management. For now, these recognition of these problems, China is currently experiments still focus on the revenue side of the preparing a government procurement law.7 budget, but better expenditure monitoring is on the agenda. Finally 12 provinces, including Citizen 's charters work. In some localities, Heilongjian, experiment with alternative ways of China has its own form of Citizen's Charters, personnel management. In the experiment, each named "Social Service Commitment System.", level of government is responsible for managing according to the system, government the administrative expenditures of the organizations and public utilities publicly state government one level below. The principle of their service commitments. Apparently, the experiment is to "manage the next level, complaints about public services went down controlling indicators, fixed amount dramatically after the introduction of the system. management, awarding with savings, and Many cities have their "mayor's hotline" on penalizing when overspent." Since the start of which citizens can express their complaints on the new management method, Heilongjian public services. apparently has had the lowest increase in administrative expenditures of all provinces. See South China Morning Post, May 5, 1997. See China Daily, October 26, 1996. 58 Enhancing Incentives for Public Sector accounting and reporting systems are far enough Performance advanced to hold units responsible for their performance. To improve results of China's public sector, the Government should enhance the incentives for Improve Information and Monitoring performance. Building on ongoing initiatives, Better treasury management. Key to improved Government should give priority to further accountability of the budget is a sound treasury differentiating between government and non- government, and between the functions of the m D system needs to momitor the proper use of various levels of government. MOF should focus sysem n eso m n ope us on accountability for results by improving govement resoue ad oime ca treasury management, accounting, and reporting. maaeet.tsol as rvd rca Prasy ike management,acce,ntand asnd ti. information for policy decisions, and for keeping Pay linked to performance, and a sound civilmagescoutb. service system, should provide government workers tile incentive to perform. In the longer run, China could adopt the Single Clarify Functions Treasury Account/General Ledger system that exists in all OECD countries, and on which the Review the Role of NPOs. China could use the IMF provides Government with advice. In fact, review of NPOs to further clarify their functions, 2,000 central spending units (Government and provide a clearer separation of Government Ministries and their subordinated NPOs) is a and non-government. As chapter 3 argued, much manageable number for a single treasury account scope exists for turning NPOs into enterprises, supported by transit accounts and integrated and dissolving remaining NPOs from the regional treasury accounts. The development of budgetary framework. In the future, non- such a system takes time, but with only minor government NPOs are likely to provide a adjustments to the accounting system, immediate growing share of government services. It is improvements could be made by: necessary for government to provide the appropriate framework, and, if required, to * Merging the treasury account and the special provide subsidies for those activities that remain fiscal account for EBFs managed by MOF. within its realm. * Merging the budgetary and the extra- budgetary spending accounts of spending From Budgeting to Performance Specification. units As a start, the budget process can be used to * Skipping one or more steps in the elaborate better specify the roles and responsibilities of procedure for requesting funds, as the only NPOs. Including output and performance binding criteria seem to be availability of indicators in the unit budgets will clarify the funds in the treasury. All OECD countries, unit's responsibilities. Over time, Government except France and Germany, have abolished could consider further steps in specifying role detailed expenditure control in and function of budgetary organizations. Some implementation, and only focus on aggregate countries require each unit to regularly submit spending limits. The budget unit itself has unit strategies that specify the aim of the unit, its the responsibility for spending according to main policies, and the key performance the budget. indicators by which achievements can be measured. Often, budgets are accompanied by Reporting on Results. China's fiscal reporting detailed business plans that cover the activities to should be expanded to include information on the be undertaken with the budget. For service results of expenditures. In line with more delivery units, a more formal system of output detailed performance specification in budgets, budgeting could over time be considered to regular reports on results, and how results replace the current system, but only if the compare with what was expected in the budget. 59 Accounting for performance. China needs to with the multi-year budget. MOF is only at the consider moving to a more useful budget start of such a system. Automation within the classification that includes an economic and an ministry is limited, and reporting systems from organizational classification and improves the spending ministries, units, and other levels of existing functional classification. A revised government have great scope for improvement. accounting system should also be able to deliver Annex 9 provides a generic outline of such a cost information for the Government's major system. programs, policies, and products. Give More Autonomy to Managers Over time, China should consider moving to an Clearer functions, better monitoring and auditing accrual-based accounting for budget units. This allows for more autonomy of budget units. worldwide trend in public sector accounting Without more autonomy, compliance with the promises to be highly conducive to performance budget may improve, but efficiency gains will orientation. Several OECD countries have only be limited. Autonomy can gradually moved to accrual accounting and budgeting, and increase by allowing more and more flexibility in many have introduced accrual elements in their . resource transfers within the operating budgets of accounting system for budget units. However, budget units. only New Zealand has introduced accrual budgeting and reporting for aggregate budget Establish detailed rules on resource transfers reports. While building such as system will take and savings. MOE should clarify the rules for time, immediate improvements could be made by virement and carryover. It could consider setting requiring spending units to report more of their l . infomatin lmit amounts under which a budget unit could grassroots information, including the information reallocate without peission from MOF. MOF that exists on economic categories, and to could also reconsider the rule that allocates redesign reporting formats and procedures such savings firsts to the budgetary reserves, because that information does not get lost in aggregation. this is a disincentive for units to save. MOF China's good records on assets that budget units could also promote efficiency in budget maintain would facilitate a possible move to implementation by making explicit arrangements accrual accounting. for dividing efficiency gains between the Value for Money Audit. The nature of both spending unit and the general budget. Currently, line a spending unit has strong incentives to spend its iteiprnan extelauds mnemehng. Ih ofull allocation in fear of losing the funding from with improving treasury management, much of it aenme.Tearneet sdi h the current internal audit could be phased out, Astralianur cos systemecouser as a and eplaed y MO's of-ste ependtu' Australian running cost system could serve as a and replaced by MOF's off-site expenditure useful example (Annex 8) monitoring. In contrast, external audits need u l e considerable strengthening, and should focus Allow more flexible staffing. Since personnel more on results than on compliance. The costs are a large chunk of the running costs of auditors of the National Audit Office need budgetary units, the scope for savings would be considerable training in this area. greatly increased, if managers were granted more Fiscal Information System. The key to a w autonomy in staffing. If managers actually have FiscaloInformatin System. the kdey tOF awell the incentive to save resources, they will rapidly functioning fiscal system that provides MOP with revise their current tendency to maximize sufficient policy information, and budget units personnel. More autonomy over personnel at the with the incentives to comply, is a versatile fiscal agency level allows for more efficient staff information system. This system should build on reductions than the current centrally-mandated an improved treasury and accounting system, reductions. should link the approved budget with budget implementation data, disbursement with procurement information, and the annual budget 60 Increase certainty of funding. Building on the China could improve procurement by requiring a proposed MTEF, China could increase the procurement plan for each project, to be certainty of funding. Once the MTEF is in place, evaluated together with a funding proposal. For and all ongoing expenditures have been large procurements, standardized procedures for evaluated on their continued usefulness, the competitive bidding should be adopted. Smaller budget allocations could be projected for several non-standard procurements could be made years. Within the MTEF, these funding levels subject to local shopping, requiring the purchaser would normally only change if policies change. to obtain at least several price quotations for the Increased certainty of funding increases the item to be purchased. China could consider a planning horizon of a budget unit, which would "Catalogue" approach for small standard benefit its operations, and would enhance procurements (Box 5.1). This would require a managers' incentives to innovate and save costs, central unit to set up competitive bidding for if at least part of the savings can be used by his items to be included in a catalogue of standard unit. items such as pens and paper. Spending units could then simply buy from the catalogue. Increase Competition More competition in public goods provisioning Actively use competitive elements in public services. China could promote the use of puts pressures on budget units to save resources. Increased competition can come from outside the competition substitutes in services not easily public sector, but could also be generated within commercialized. Measures Government could the public sector, by increasing cost recovery and consider include: publicity on performance. * Published service standards. Building on the Involve the non-state sector. Government should local initiatives with service standards, consider actively encouraging the emergence of a Government could require those standards nonstate service delivery that would compete for for each spending unit. Government could clients with NPOs. This would require the use standards not only for services to the elimination of general price subsidies for public public, but also for internal processes. services, the monetization of service delivery in rublishing the standards widely, and the public sector and the targeting of poor monitoring a unit's performance will force beneficiaries who cannot pay market prices and unit personnel to meet those standards. fees. A major benefit of private sector * Benchmarking. China's decentralized development in service delivery would be the government lends itself well to the use of creation of an alternate market for surplus NPO benchmarking. Comparing the performance staff. of similar institutions in various regions, and publishing the results of these comparisons, Strengthen cost recovery in NPOs. Increasing can be a powerful tool for enhancing cost recovery for public services can be a performance. powerful tool for efficiency. This would include * User funding. Shifting funding from reducing the list of NPOs whose staff is paid by provider to user could increase competition the budget and accelerating the phasing out of and efficiency within the public sector. For grants. Combined with competition by the example, funds for training could be private sector, NPOs would be impelled to allocated to the operational departments improve services while reducing costs. NPOs rather than to the training center directly. that can be fully financed from user charges Over time, this philosophy could be applied could be commercialized, - and divested from to policy funding as well, e.g., by providing Government. vouchers for education instead of funding Improve government procurement. China's plans schools. to improve public procurement are a priority. 61 Motivate Civil Servants Revamp Intergovernmental Fiscal Relations Better incentives for budgetary units may China needs to develop a pay system in the public administration that will provide rewards increase public sector performance at each level for strong performers. The job evaluation and of government. But to eliminate the duplication n t is of functions among levels of government, and the grading that the Minstry of Personnel i undertaking is a first priority for this. Job glaring disparities among provinces, grading should link civil service jobs with intergovernmental fiscal relations need Box 5.1: US and World Bank Procurement Rules US Federal Government Procurement Public procurement by the US Federal Government is done by several agencies. The General Service Administration (GSA) does procurement of some items, mostly large-volume purchases, mainly for domestic purchases. The Federal Supply Service (FSS) buys, stores, and distributes a variety of goods for GSA. Some other agencies, including the Department of Defense also has procurement authority. GSA publishes a Govemment Supply Catalogue. For individual or groups of items in the Catalogue, GSA organizes competitive bids The winner gets listed in the Catalogue, which lists complete item descriptions, prices, the name of the winning bidder, and the period of bi validity. Most purchases up to an amount of US$25,000 are done through the Catalogue. Construction, Goods, and Commodities. According to the Federal Acquisition Regulations, all US federal government contracts costin] more than US$25,000 must be awarded by competitive bids under the terms of the Competition and Contracting Act (CICA) of 1994. The Ac requires full and open competition by sealed bid. All upcoming bids are advertised in the Commerce Business daily, a publication of the US Commerce Department. There are exceptions to competitive bidding, as listed in the CICA, including criteria when "sole sourcing" is allowed. Each exception to competitive contracting must be justified in writing. Consultants. Procurement of consultants follows a procedure by which negotiations take place with bidders. A request for proposal i published in the Commerce Business Daily, and proposals are reviewed on the basis of technical and financial proposals announced in advance Those firms that are in a "competitive range" are invited for negotiations. World Bank Procurement Rules The World Bank has a set of rules for procurement similar to those of the US government for goods and services procured under its loan and credits. The competitive processes for goods are split into International and National Competitive Bidding (ICB and NCB). ICB is used fol large procurement, where "large" is defined per country and per category of goods. Usually, the threshold is larger for construction. The competitive bid can have a one stage or a two-stage process. The one stage is mostly used for straightforward procurements, and is similar to the US system. The two-stage bid allows for discussions with the suppliers that meet a certain minimum technical score on the criteria set in advanced. These bidders are invited for discussions on the technical proposals. The criteria for the technical proposal can be changed by tht buyer during that stage, and the bidders can adjust their technical bid. The reason for this is that the buyer can benefit from the information in tht technical bids, and can adjust its technical criteria accordingly. In the second stage, bidders submit a revised technical proposal, and a financial proposal. A third technique under Bank procurement is international or national shopping. For smaller, fairly standardized goods, a buyer car obtain several price quotations, either internationally or nationally, and can award the contract to the lowest bidder. A recent addition to Bank rules excludes suppliers who have been found to violate competitive practices in earlier procurements, e.g. b) bribing officials from bidding. equivalent ones in the market economy, where revamping. A full treatment goes beyond the pay reflects productivity. Increasingly, the proper scope of this report, but China could consider benchmark will not be the State enterprises, but further clarifying expenditure assignments, the non-state sector. However, providing similar accelerate the introduction of an equalizing pay for similar jobs is not enough: it is grants scheme, and make selective use of performance that deserves rewards. For this, the earmarked grants to promote national goals. civil service needs a system to evaluate performance on a regular basis, and to link wage Clarify Expenditure Assignments. Reducing increases and bonuses to performance. China overlap among levels of government remains a could over time consider delinking core civil key area for further efficiency gains. Annex 10 service wages from NPO wages. provides some general principles and international evidence on expenditure 62 assignment. No single right expenditure equalization. Although China could choose assignment exists, but the assignments should be different means to bring greater equalization, clear to the participants in the budget process to without it, the country will not improve service avoid costly overlap. As a start for a more delivery in poorer provinces. comprehensive review of assignments, Government could include a clearer division of Earmarked grants. In addition to general functions, expenditures, and revenues in every equalization grants, China could rethink its new policy proposal and law. system of earmarked grants. The current transfers are mainly used for SOE subsidies, but Revise Intergovernmental Fiscal Relations. they could become a tool for furthering national China will not overcome the regional disparities goals, without imposing unfunded mandates on in service delivery without further revision of lower level governments. Since most of the Box 5.2: The Four Pillars of Successful Civil Service Systems Working environment. The ability of economic technocrats to formulate and implement policies in keeping with politically formulated national goals depends on the extent to which they are effectively insulated from lobbying for special favors by politicians and interest groups. Various systems exist to insulate technocracies: Hong Kong, Japan and South Korea have evolved administrative and legislative systems which give primacy to, the bureaucracy in proposing laws. In Japan, the bureaucracy is further protected from political pressure by the National Personnel Authority, an independent body that sets the bureaucracy's pay scales and promotion policies, administers civil service exams, and makes most appointments. Japan's Prime Minister names only his ministers and, except in a few cases, one of the two vice-ministers in each ministry. The NPA is responsible for all other appointments. Merit-hased recruitment and promotion. Highly qualified staff will be attracted to a civil service career if entry and promotion are directly related to their ability to compete. In successful civil service systems such as Japan and Korea, recruitment revolves around highly competitive entrance exams, administered by national personnel authorities. Success rates are low, which indicates the high standards expected by applicants. Promotions in Japan are based on a combination of seniority and a host of performance indicators that differ across ministries. Because law fixes the number of personnel in the bureaucracy, competition for promotions is intense. Furthermore, throughout the career of a civil servant, the lack of career progression points to inadequate performance and will be sanctioned by early dismissal. Incentive-based compensation. In bureaucracies as in all other walks of life, you get what you pay for. Most successful bureaucracies have put into place incentive systems that are geared to make the civil service competitive with jobs offered in the private sector at most ranks. Obviously, senior civil servants may not receive compensation equivalent to Chief Executive Officers in the private sector, but at those levels in bureaucracies, other intangible incentives are present. Thus, monetary compensation is a very powerful incentive and the divergence between compensation in the civil service and the private sector, at all hierarchical levels, should not be excessive. Predictability of career paths. A final requirement for a successful bureaucracy is a well-defined, competitive career path, with a substantial prize for those who make it to the top. Young entrants into the civil service will be willing to commit themselves to a civil service career if there are reasonable prospects that sustained high performance will allow them to enjoy relatively rapid promotions and to achieve a high rank in the administration. In Japan as in other East Asian civil service systems, retirement comes early, and the rewards to a successful bureaucrat are substantial, extending beyond the pay, perks and prestige to include golden parachutes. intergovernmental fiscal relations. The 1994 service delivery takes place at subprovincial reforms insufficiently redistributed resources levels, central government could consider among provinces, and will continue to do so for a targeting these grants to municipal or even long time. Annex 11 suggests an county levels. Although the effective design and intergovernmental grants system that will more monitoring of earmarked grants is no easy quickly reduce horizontal disparities, by availing matter, the task will become less cumbersome an increasing share of the tax return grants for with improved budgeting, treasury management and accounting systems. 63 The Changing Role of MOF no longer need to deal with detailed budget negotiations to cut expenditures within available MOF can play a major role in the reforms means. proposed in Chapters 3-6 of this report. In many OECD countries, MOF has taken the lead in In budget implementation, MOF's role in expenditure management reforms, by generating expenditure monitoring will increase, once a discussion on public sector performance, proper treasury management system is in place. designing new budgeting procedures, reporting A good treasury system would do away with formats, accounting procedures and developing much of the on-sight internal audit of the evaluation capacity. MOF could be this catalyst Supervision Department, and replace this by off- of change for China. site monitoring. However, with increasing autonomy of line ministries and agencies, But what would MOF's role be in a reformed expenditure monitoring will become more and expenditure management system? Much more a control of aggregates, not a detailed depends on how the budget process evolves. If scrutinizing of budget requests. the budget becomes the Government's main tool for policy, MOF is likely to become more With the changing role of MOF, its organization involved in the analysis and evaluation of policy. will change as well. The current structure-with As a minimum, MOF would be involved in over 20 departments-can over time be greatly setting the standards for policy evaluation, and be simplified. MOF could consider a departmental in charge of determining the budgetary structure focused on its core processes, namely: implications of new policies. Whether it will play budget preparation, budget implementation a major role in setting the policy agenda depends (treasury), internal audit, and tax policy (Figure to a large extent on the role of SPC, which 5.3). A separate department could then take care currently determines much of this agenda. If the of all administrative services. In this suggested Government decides to move to a multi-year structure, the sectoral departments would be budgeting framework, MOF is the obvious abolished. Their functions in budget preparation candidate for managing the forward estimates, would be included in the central budget office, and determining the budgetary room available for whereas their budget implementation function new policy. would become part of the treasury office. The central budget department would be in charge of But improved links between strategic policy and the annual budget, the multi-year expenditure the budget will greatly reduce MOF's role in framework, and overall fiscal policy. The detailed budget preparation. The expenditure treasury department could assume much of the review committee, or some high level political tasks in budget implementation, including cash body, will set overall policies and sectoral management, debt management, and accounting. expenditures within which the line ministries make their trade-off. In principle, MOF would 64 Figure 5.3: Possible structure of MOF Finance Minister Deputy Ministers Internal Audit Central Budget Central Treasury Tax Policy Administrative Department Department Services Fiscal Policy Cash and Debt Management Budget Affairs Bde execution Evaluation Acutn Systems development Asset management 65 ANNEX ] ANNEX 1: CIINA'S EXPENDITURE MANAGEMENT SYSTEM Legal Framework In March 1994, the NPC approved an organic Budget Law, which became effective January 1, 1995. The law replaced the State Council "Regulations on State Budget Management" of October 1991. The Budget Law specifies the budgets of the various levels of government; the roles and responsibilities of the agents in the budgetary process; the scope of the budget; the various steps in the budget process; and the legal responsibilities of the agencies involved in the budget process. The law delegates issue of detailed regulations to the State Council. Such regulations were issued on November 22, 1995 (State Council Decree No. 186). Further legal underpinnings of the budget process are: MOF regulations on "General Budgetary Accounting System of Fiscal Departments" and the "Budgetary Accounting System of Administrations and Institutions" effective January 1, 1989; and the Sate Council Circular No. 29 of July 16, 1996 "The Decisions on Strengthening the Management for Extra-budgetary Funds." The Budget Law (Art. 47) states that each level of government with a budget institutes a treasury and that the central treasury is managed by the PBC. The State Council issued regulations on Treasury Management on July 27, 1985, replacing regulations of 1950. MOF and PBC issued detailed regulations on December 13, 1989. In practice, MOF is primarily responsible for drafting regulations concerning the Treasury. Budget Scope Chapter III of the Budget Law defines the scope of the budget. It defines four revenue sources of which tax revenues and revenues from State-owned assets are the most important. It further defines, in general terms, six expenditure categories, among which are economic construction and administrative expenditures. China has five levels of budgets (central, provincial, municipal/prefectural, county, and town/township); each approved by their corresponding levels of People's Congresses. Not all towns and townships present a budget. Since effectiveness of the Budget Law, the local People's congresses approve the local budgets. They check whether the budget conforms to national laws and local regulations. The local budgets are submitted to MOF for the record and for compilation of the State budget. The budget speech of the Minister of Finance to the NPC still presents the consolidated State budget, but the NPC only approves the central budget. The budget format is the "Dual Budget System" which divides revenue and expenditures in a recurrent and a construction budget. The recurrent budget contains investment expenditures in "nonproductive" sectors such as education and health, and the construction budget contains recurrent expenditures to support productive activities, such as geological prospecting. The budget expenditure classification is a mixture of economic classification (e.g. capital construction) and functional classification (e.g. education, science and Health). However, even the functional classification is a mix of functional and organizational classification, because the accounting system does not delineate separate functions of spending units clearly. Borrowing and repayment is accounted for "below the line" since 1997, but so are interest payments. 66 ANNEX 1 Extra-budgetary Funds A substantial amount of government resources falls outside the scope of the budget. These "extra-budgetary funds" of government agencies and institutions consist of levies, fees and tax surcharges that accrue mainly to local governments. There are as many as 600 types of funds, not all approved by the relevant authorities, which amounted to Y380 billion in 1995. The EBFs originated in times of strict central budgetary planning as a means of providing some limited autonomy to government bureaus and spending units. By 1995, however, these funds had grown to over 5 percent of GNP, or almost a third of government resources. MOFs' Comprehensive Department compiles statistics on EBFs, and the State Council has issued various regulations on them, including Decision No. 29, which orders integration of 13 funds amounting to Y1 15 billion in the budget, and tightens controls over other EBFs (Annex 5). The State Council circular abolished the off-budget nature of the "self-raised funds" which lower levels of government raised. This category, which was not registered in the fiscal reporting system, consisted mainly of levies on township and Village Enterprises at County and Township levels. Quasi-Fiscal Resources The banking system has provided financing for government expenditures outside the budget. "Policy loans" extended by the State banks, and in large part refinanced by the PBC, added an estimated 5-7 percent of GDP to the government resource envelope. Policy loans provided an interest rate subsidy to the recipient of 10-70 percent of standard interest rates. These subsidies were in principle to be covered by the Budget, but banks regularly report on arrears of the Budget. Since the establishment of the Policy Banks in 1994, policy lending through the commercial banking system has been reduced. The State Development Bank provides about Y1OO billion per year in loans to the government's priority projects. Policy loans are directed according to the credit plan, which is drafted by the State Planning Commission (SPC) and implemented by the PBC. Beneficiaries of these loans are enterprises in sectors that implement public investment projects, or that make losses due to the social burdens and price limitations imposed on them by government. The bad debts in the banking system, a contingent liability for the budget, are estimated to amount to 20 percent of the banking loans, or almost 20 percent of GDP. Budget Year The budget year coincides with the calendar year, January 1-December 31. Since budget approval only takes place in March at the central level, the budget year is effectively reduced to nine months. According to the Budget Law, Article 43, until the budget for the current year is approved, expenditures not exceeding those of the previous year are allowed. In practice, this may mean that some local spending units do not have their budgets approved until May of the current fiscal year. 67 ANNEX 1 The Budget Process Budget preparation. Formally, the State Council is responsible for drafting the budget, but the Minister of Finance is assigned responsibility for practical organization of the work and compilation of the draft budget. Budget preparation starts in October-November with the State Council's letter to central government ministries and provincial governments on the expected macroeconomic developments for the next fiscal year (see Table Al.1). The letter, drafted by MOF contains the expected revenue and expenditure developments. These figures are based on historical experience, the current year's budget implementation, and on SPC's letter on macroeconomic developments. The State Council letter does not provide expenditure limits, either overall, sectoral or organizational. The ninth Five-Year Plan, however, has as a goal to "basically eliminate" the budget deficit. Table A1.1: Indicative Budget Calendar October-November State Council orders drafting of the budget and sets out broad guidance for the draft October-December Spending Units, Ministries, and local governments compile their draft budgets, and submit to the sectoral departments in MOF (in localities, some proposals do not come in until February of the budget year) November MOF consults with Ministries and local governments on the next year's budget November-December National Finance Conference; MOF drafts the Central budget and compiles the State budget . January MOF submits the draft Central and State budget to the State Council. Discussion in the State Council on outstanding issues and major projects January-March Audit of Spending units' accounts by the National Audit Office February Presentation of the draft Budget to the Standing Committee of Finance and Economics of the NPC March Presentation of the Central and State Budget and the preliminary Final Accounts by the Finance Minister to the NPC March-May MOF passes on the approved budgets to the spending ministries. Spending ministries pass on the approved budgets to the spending units January-April National Audit Office audits the Central budget June-July Presentation of Final Accounts by the Finance Minister to the Standing Committee of the NPC Source: MOF, Budget Law. Spending units and line ministries prepare their budget proposals on the basis of: (1) laws and regulations; (2) the State Council's circular on the budget, which prescribes the budget to be prepared in line with (a) the growth rate of GDP, (b) forecasted revenues and (c) the social and economic development plan; (3) the quota and standard expenditures of every spending unit; and 68 ANNEX 1 (4) the results of the current year budget implementation. In practice, this has led to an incremental budgeting which took last year's expenditures as the base of the budget, with an increment related to the overall revenue increase (Base number method). Enterprises do not submit a budget but a financial plan, which may include budget support for the enterprise. The ministries review the proposals and consolidate them with the budgets for ministerial expenditures. The ministries submit the consolidated proposals to MOF, together with about 30 pages of motivation, summarizing the proposals of the spending units. The budget submission is summarized in several standard forms. The mission was handed a proposal that contained three forms. Spending for a Ministry is divided into main categories, each subdivided in: (1) personnel costs, (2) office expenditures and (3) special projects. Various expenditure norms for personnel and office costs exist, which are included in MOF's budget circulars. The budget proposals take the EBFs of the spending units into account, but they are not budgeted for. The 13 EBFs that the State Council ordered integrated in the budget (including the Electricity Fund and the Railway Fund) are gradually integrated, although not all localities are yet complying with this regulation. Apparently, a ministry can submit several budget proposals to MOF. The proposal for budgetary money to enterprises subordinated to the ministry is separated from that of the budget for Non-Profit Organizations under the ministries. The reason seems to be that separate sectoral departments in MOF are responsible for reviewing the drafts. For example, the Ministry of Agriculture submits two drafts to the Agriculture Department in MOF, one for the NPO division, and one for the Enterprise division. Sectoral departments within MOF, in consultation with the Budget Department, review the proposals. The review focuses on additional spending over last year's expenditures. Usually, a spending ministry would end up with about 20-30 percent of the requested additional budget. The overdemand is used as a negotiation strategy: it shows the priority of the sector and positions certain proposals for the following year's negotiation process, because a proposal is usually accepted after it has been submitted for a number of years. The submitted budget usually undergoes several rounds of written revisions. The sectoral departments compile the proposals of various ministries into a sectoral proposal, which is passed on to the Budget Department. For instance, the Agricultural Department in MOF compiles the budgets for the Ministries of Agriculture, Forestry and Water Resources and the Bureau for Weather Forecast into a budget for "Agriculture." The compiled sectoral budgets are submitted to the budget department of MOF. The budget department compiles the Central budget and aggregates the local budget estimates to the Central budget to compile the State budget (for general government). The local government budget cycle starts considerably later than the Central budget cycle, in some localities as late as March. Local bureaus submit their budget estimates to the local budget department of the finance bureaus, which, after negotiation, pass on the budget to the Mayor's meeting, in which outstanding issues are settled. In the localities visited by the mission, the original budget draft of the line bureau was submitted for the Mayor's meeting, together with the finance bureau's comments. The Mayor's meeting should come to agreement on outstanding issues, before submitting the draft to the Local People's Congress. 69 ANNEX 1 Zero-Based Budgeting. To move away from the incremental base number technique, several local governments are experimenting with ZBB since 1993. ZBB puts more emphasis on the planned activities of a spending bureau rather than on last year's budget. In principle all expenditure proposals are reviewed, not just the increment over last year's budget. Expenditure norms for office expenses are included in the finance bureau's budget circular. The spending bureaus are required to put considerable detail on its "special projects" in the budget draft, which allows the finance bureau to probe the proposed expenditures more thoroughly, and to better assess the expected outputs of the bureau. According to the local authorities, their budgeting practice also puts more emphasis on controlling personnel costs as the more detailed budget proposals allow for a check of actual personnel rather than the quota of personnel assigned by the Personnel Bureau. The local authorities state that the ZBB technique has enabled them to better control overall expenditures, control wage costs and readjust expenditures in priority areas. However, they feel that the technique has its limits, because a large part of the budgets consists of personnel costs ("meal finance") which cannot be altered through the budget process. Macroeconomic Coordination. During the fall, MOF, SPC and the PBC consult with each other on the budget plans and on the budgetary contribution to the investment plan. MOF negotiates with ministries on their budgets and with local governments on the central contributions to the local budgets and the local contributions to the central budget. In November/December, the National Finance and National Planning Conferences finalize the investment plans and budget proposals of ministries and localities. MOF compiles the budget results, and presents this to the State Council for review and approval. Although the MOF has started a five-year rolling budget plan, there seems to be no coordination with SPC's five-year rolling investment plan. Recurrent costs for government investment projects are determined at the end of the investment construction, and only then taken into account in the budget. Budget approval. After State Council review and approval, the "main contents" of the draft budget are submitted in February/March to the Standing Finance and Economics Committee of the NPC for detailed review. In March, the Minister of Finance presents the budget speech to the People's congress, and the Standing Committee presents a report on the draft budget. The NPC approves the budget on the basis of the speech and the report. The budget has legal status after NPC approval. After approval of the Central budget, MOF informs the ministries on the approved budget. In turn, the ministries inform their units on their budgets. Usually, by the end of April, all central units know their budget. At the local level, this may take until end of May. Extra-budgetary Plan. State Council circular No. 29 orders an annual extra-budgetary plan to be prepared, but in practice this is not yet done, except for the 13 funds that are now included in the budget. MOF issued the regulations "Notice on formulating the Plan for Extra- budgetary Revenues and Expenditures" in early 1997. Budget Adjustment. Budgets can be adjusted by (a) a budget adjustment passed by the Standing Committee on Finance and Economics of the NPC and (b) adjustments made by Government. The Budget Law gives no clear guidance as to the circumstances in which either method may be used, but in practice only method (b) is used. Budgets are adjusted regularly, and spending units and ministries continue to propose additional spending measures during the year. The reason for this is that revenues are always larger than budgeted. Criteria for approval of additional expenditures are unclear and seem to blur the original intention of the budget to some 70 ANNEX I extent, as the expenditure distribution in the final accounts differ significantly from those in the budget. Payment Process Budgetary Expenditures. Together with the budgets, monthly or quarterly budget plans for the spending ministries are approved. These plans serve as guidance for disbursements from the Treasury. Disbursements take place in two ways: (a) "general use" method and (b) "quota" method. Personnel and office expenses are usually governed by the quota method; special purpose expenditures by the general use method. General use method * The spending units submit a request for funds to their supervising ministries. Spending units have their own bank accounts, but they do not submit a copy of their bank statement together with the request for funds. The ministry consolidates the spending units' requests and submits a request for funds to MOF. * The relevant MOF sectoral department scrutinizes the request of the ministry and hands it to the MOF budget department, accounting division. The accounting division in the budget department issues a payment order to the treasury department of PBC. * The PBC National Treasury Department (NTD) fills out a payment form and submits that to the accounting department of PBC, which then authorizes transfer from the treasury account. The authorized payment form is brought to the Beijing branch of the PBC, which submits it to the clearinghouse. Funds are then transferred to the accounts of the relevant ministries in the Specialized Banks. * The ministry transfers the funds to the spending units' account in a state bank. The spending units then make the payments from their bank accounts. Quota method * The spending units submit every quarter a quota request to the supervising ministry, which submits a consolidated request for the quarter's quota to MOF. * After review by the sectoral department, MOF budget department notifies the NTD, which then issues a quota form to the spending ministry and its bank. The quota is regarded as a letter of credit by the bank, which then entertains requests from the spending ministry for allocation or actual expenditure against the quota. No commission is paid to the bank. * At the end of each day, the bank clears the balance of the spending units at the local PBC branch, MOF transfers a "refund order" to the treasury, which differs from the payment order, but is treated in the same way. * The capital construction department of MOF approves budgetary expenditures for capital construction. After approval, accounting division of MOF's budget department releases the payment order. The payment order is sent to the China Construction Bank. At the end of the month, the banks report to NTD on their disbursements against the quota. The NTD requests MOF for a reimbursement of the quota account. 71 ANNEX 1 NTD reported that the quota method was taken from the Soviet system in the 1950s. The credit supplied by the banks through this method was said to compensate for the treasury money that was in commercial banks at any point in time. In 1996, about Y20 billion or less than 4 percent of total expenditures was made by quota. Subsidies for lossmaking enterprises are not treated as expenditures but as refunds on taxes, both in the budgetary accounts and in the expenditure process. The enterprise applies for such subsidies through the supervising ministry, which submits a consolidated request to MOE. Upon receipt of the payment order, the Construction Bank transfers money to the accounts of spending ministries, which transfer the money to the bank accounts of the spending units. The account in the Construction Bank is replenished from the Treasury account by bills of exchange approved by MOF. Certain budgetary interest rate subsidies are also authorized by the Capital Construction department, but are transferred from the treasury to the State Development Bank (SDB) which has taken over part of the policy lending from the State Banks. The SDB uses the People's Construction Bank of China as a disbursement agent. Tax refunds (e.g. for value added tax paid on exported goods) are applied for at the tax bureau, which reviews the application, and after approval sends the application to MOF or the finance bureau. MOF or finance bureau issues a "refund order" to the treasury, which transfers the money to the enterprise or, where applicable, to the ministry that made the application. The new system of credit invoice VAT, combined with a surge in exports since 1994, has caused some cash flow problems for the treasury.' To counter this problem, refunds were previously paid in installments. On July 1, 1995, the crediting of invoices was restricted to 14 percent, instead of the 17 percent VAT rate; since January 1, the refund rate is further reduced to 9 percent. Extra-budgetary Funds. EBFs are disbursed by a different procedure than that for budgetary funds. Since no budget for EBFs is passed, the use of the funds is determined during the budget year. Extra-budgetary revenues are now kept in a "special fiscal account" by the finance bureau. This differs from the practice before State Council circular No. 29, when the funds were kept in a "special account." The difference is that EBFs are now no longer considered as owned by the collecting unit. The spending unit only has a transit account for extra-budgetary receipts, and an account for extra-budgetary expenditures. All extra-budgetary accounts, including the one of the finance bureau/MOF is held in a commercial bank. Each spending unit submits a request for the disbursement of EBFs to its supervising agency. After review and approval, this is passed on to the finance bureau/MOF. The units request either monthly, quarterly, or ad-hoc funds. Monthly or quarterly requests for funds usually concern wages or office costs. Ad-hoc requests are made for special projects. The request form has three copies: one for the applicant, one for the supervising agency and one for the finance bureau. After approval, the finance bureau issues a payment form, which it passes on to its bank. The form has 4 copies: two for the finance bureau-one before and one after the funds transfer-and one each for the bank of the finance bureau and for the bank of the spending unit. 1 In 1994, refunds were about 50 percent higher than in 1993. 72 ANNEX] After approval, approved funds are transferred from the Finance Bureau's special account to the spending unit's extra-budgetary expenditure account from which payments are made. Reporting. The central treasury reports daily, monthly and annually to MOF, both on its own operation and on the operations of treasuries at provincial level. Daily and monthly reports consist of aggregate numbers on revenues, expenditures, and treasury account balances. Over ten expenditure items and the main revenue items are reported. Annual reports provide detailed reports on all revenue and expenditure items. Lower level treasuries report both to the finance bureau at the same level and to the treasury one level above. Once revenues enter the county treasury, it takes about eight days before the revenues are reported to the central treasury, if the county is not computerized. About 25 percent of the counties are computerized. Spending units report monthly and annually to supervising ministries, which compile reports for MOF. MOF, however, only substantially evaluates the annual reports. The tax administration reports daily, monthly and annually to the finance bureaus at the same level and to tax bureaus one level above. Final accounts. At the March session of the NPC, the Finance Minister presents preliminary final accounts for the previous year. The final accounts are presented to the Standing Committee of the NPC in June/July. Final accounts are compiled by MOF, on the basis of the final accounts drafted by the spending ministries and the final accounts of the NTD. The spending units and ministries submit a bank account statement together with their final accounts. Audit Authorities of the People's Republic of China audit the final accounts. 73 ANNEX 2 ANNEX 2: EXPENDITURE MANAGEMENT ISSUES RAISED IN BUDGET SPEECHES 1992-1997 Year Issue 1992 * Administrative expenses much more than budgeted * Budget categories do not reflect priorities * Appropriate control over future bonds issue * Give more attention to financial work * Raise efficiency * Finance losses in excess of the contracted amount * Tax exemptions * Smooth sale of treasury bonds * Effective control over expenditures * Arbitrary extra-budgetary expenditures * Clean government * Careful audit and supervision * Combat formalism, reduce number of meetings * Reduce number of celebrations * Reduce expenditures on excess staff * Control soaring health care costs by shifting responsibility to health institutions * Purchase of new equipment by departments * Increase in discipline of state finances 1993 * Reviewing and rectifying the scope of expenditures and improving the management of finances * Some localities not able to meet normal payroll * Launching of too many projects * Overstaffing, causing total expenditures for personnel to be excessive * Macroeconomic control needs to be improved and strengthened * Leaders in localities should pay more attention to financial work, improve management expertise of finance, tax workers * Find new ways to efficiently manage funds for supporting agriculture * Unplanned development and redundant construction furthered by inappropriate administrative interference and preferential policies 74 ANNEX 2 * Control over expenditures, putting stop to extravagance and waste * Strictly enforce the system for budget adjustment * Reduce number of meetings and documents * Institutions should improve the management of their funds, incorporating both funds allocated by financial departments and their earnings into a unified plan for revenues and expenditures (24) * Exercise strict control over extra-budgetary funds (24) * Scientifically define the state's function in the management of society (24) * Give full play to the supervisory function of the finance departments * Review all finance and taxation laws 1994 * Expenditures exceeded budgeted amount by big margin * Starting projects at random * Spending administrative funds on unnecessary new and foreign items * Extravagance and waste * Acute problem with supply and demand of funds * Some localities not able to meet their normal payroll * Making management of finances more standardized, better legal framework * High expenditure base for 1993 made it necessary to increase essential expenditures spent on agriculture, education, science, technology and other key areas. * Reductions in revenue, increases in expenditure, compliance with budget * Legal footing of financial management * Reform method of issuing treasury bonds * Controlling overall expenditure, funding for key projects * No new project launches * Abandon projects without funding or with unclear future * Wasted funding * Travel under pretext of holding meetings * Wine and dine * Deepen reform of financial management in public institutions (29) * Proceeds from land use, sale of public housing and other form of revenue which should come under budgetary control should come under budgetary control (29) 1995 * Expenditures exceeded the budgeted figure by a fairly large amount * Problem of weak financial budget restrictions still remains unsolved (p5) * Losses and wasting of revenues are still very common. (6) * No big adjustments have been made in the original pattern of distribution of interest (6) 75 ANNEX 2 * Our macro-regulating capacity is clearly insufficient (6) * Some localities were not even able to pay wage increases or even guarantee the normal payroll * Because all fields of endeavor have, in recent years, been calling for increased expenditures to accelerate development financial expenditures have stayed 2 to 4 percent higher than the increase in revenues (11) * There is great potential for cutting down expenditures * Local governments should pay close attention to adjusting the pattern of utilization of funds for agriculture and concentrate on supporting agricultural infrastructure facilities. * Poor performance of SOEs * Separate functions of government from those of enterprises * Guarantee funding for key projects while cutting back on general spending * Continue to increase investment in such key areas as agriculture, education, science and technology * High priority to meet the payroll and carry out wage reform measures introduced last year * Waste is due to inaccurate feasibility studies, factors related to price rises, improper management of construction funds and losses in the process of project subcontracting (22) * Tighten control of institutional spending through financial management. 1996 * Budget overruns were in evidence only in the local budgets (6) * Some central budget expenditure came from local governments in implementation of the budget, thus increasing actual local spending (6) * We introduced a system whereby central government checked on quotas for administrative expenditures an the provincial level * Some regions implemented a zero-based budgeting system * Central finance still has problems * An effective solution has not yet been found to the problems of the transformation of budgetary into extra-budgetary funds (7) * Expenditures still grew too rapidly (7) * The problem of back wages owed to office staff and workers in administrative departments and institutions still exists in some localities * Diverting funds for agriculture to other purposes * Straighten out the distribution relationships (19) * Operation of the revenue sharing system at and below the provincial level * Some local governments have converted budgetary funds into extra-budgetary funds without authorization to evade budgetary management and supervision. The common practice of establishing private funds has become a source of indiscriminate 76 ANNEX 2 distribution of bonuses, extravagance and squandering (22) * These problems have not only become a great drain on the state tax revenue and assets, disrupted fair competition and market order and adversely affected the effective implementation of state macro-control policies, but also corrupted some cadres (22) * We should improve the management of extra-budgetary funds. All items that should be subject to budgetary control according to the decree of the State Council must be incorporated into the budget (22) * Concentrate on uncovering such problems as keeping two accounts, tax evasion, conversion of budgetary funds into extra-budgetary funds and unauthorized private funds (23) * We shall carry out overall checks on the use of extra-budgetary funds (23) 1997 * The deficit in the central budget is still too large (6) * Because expenditures are not effectively controlled, they increase at a high rate (6) * Extravagance and waste are prevalent (6) * The unauthorized establishment of funds and the collection of arbitrary charges is still prevalent (6) * The diversion of state financial resources has not been checked (7) * Prevent extensive investment and avoid rushing headlong into action without considering feasibility (14) * Step up our efforts to cut down the number of meetings, strictly control and approve the traveling expenses for Party and government officials who go abroad and resolutely put a stop to the unhealthy practices of holding banquets and going sight- seeing at public expenses (16). 77 ANNEX 2 Stated Goals in Budget Speeches 1992-1997 Year Goal 1992 * Financing of economic development * Fight natural disaster * Support Reforms in all fields * Increase investment in agriculture, science, technology, education and other key sectors * Invigorate large to medium SOEs (LMSOEs) * Appropriate subsidies to staff of administrative departments and government institutions for grain price rise * Continue to aid poor regions * Clean up chain debt 1993 * Increase investment in agriculture * Invigorate LMSOEs * Increase investment in science, technology and education * Stress key sectors * Economic development is central task * Decrease of unprofitable enterprises in the budget 1994 * Continue to strengthen and improve macro-regulation * Promote overall development of the rural economy * Improve operation of LMSOEs * Increase investment in science, technology and other key areas * Improve macro-control and promote rapid and sustained development 1995 * Further adjust the pattern of expenditures * Increase expenditures for further building up the national strength * Practicing strict economy and build up the country through thrift and hard work, make overall plans and take all factors into consideration and ensure the construction of key projects (7) * Curbing inflation and reducing the excessively high price index * Continue to ensure the grain risk fund and the non-staple food risk fund (11) * Increase subsidies for minority areas (18) * Paying special attention to the quality and efficiency of economic growth (19) 78 ANNEX 2 1996 * Adopt a relatively stringent fiscal policy, support economic development, continue to improve the reform of finance and taxation, work to rectify the financial and economic order, tighten control over expenditures, oppose all forms of extravagance and waste and strive to reduce the deficit (8) * Basically eliminate the deficit over the Ninth Five-Year Plan (12) * Ensure adequate investment in key projects * Increased income should be mainly used to solve the problem of overdue wages, ensuring key expenses and paying off past financial debt * Tighten control over use of funds and ensure better results for expenditure (12) * In education, science and technology, more money should be spent on improving and updating existing teaching and research facilities, and not on launching new projects (13) * Do all we can to prevent the diversion of funds earmarked for key projects to any other purpose (13) * Try and spend less money doing more (13) * In the management of expenditures we should continue to strictly carry out a relatively stringent financial policy, readjust the structure of expenditure and make sure that the expenditures that should be cut are cut without fail. We should thoroughly fight corruption to put an end to evil practices in the use of public funds such as extravagance, waste and ostentatious behavior and tighten the control over expenditures (21) * In managing expenditures, we should work to review and spread useful experiences and methods for reducing expenditures (21) 1997 * Make support of agriculture a priority in our financial work (13) * Promote the two fundamental shifts in the economic system and the mode of economic growth (14) * Priority to reducing administrative expenditures (15) * Collect the land use receipts which have not been turned over to the state finance (15) * The increase in expenditure must be lower than the revenue (15) * The pattern of expenditures should be readjusted (15) * We must strictly control the size of staff in state organs * Stop providing administrative funds to some units which have ceased to perform the functions of administration (16). 79 ANNEX 3 ANNEx 3: EXPENDITURE NEEDS FOR PRIORITY AREAS Underfunding of priority activities, including infrastructure, the environment, health and education services and poverty alleviation, which perpetuate unacceptably poor living conditions for a large segment of the population, may threaten the sustainability of growth. This annex presents some estimates of the extent to which government budgetary expenditures in the social sectors and in poverty alleviation fall short of needs. Education and health China's record in human capital formation has been impressive and its social indicators are high in comparison with low-income countries, approaching those in higher middle-income countries. However, recent years show a declining trend in social sector spending, eroding China's favorable standing among comparable Asian countries. Public expenditures on education declined from 3.1 percent of gross domestic product in 1985 to 2.4 percent in 1994 and again to 2.05 percent in 1995. This is in contrast to the Asian NICs, where steady increases in public spending on education (especially basic education) played a fundamental role in sustaining their high growth. And relative to its East Asian neighbors, China's spending has been decreasing since 1985, and now is the lowest in the region (see Table A3.1). Part of China's declining share of education expenditures in gross domestic product may be explained by demographics and China's one-child policy. Also, fees and self-raised funds have increasingly supplemented budgetary resources, raising the burden on poor households and reducing their access to schools. Table A3.1: Public Expendituresa on Education (Percent of GNP) 1985 1989 1992 Hong Kong 2.8 2.8 2.9 Korea, Rep. of 3.0 3.6 4.2 Singapore 5.0 3.4 4.4 Malaysia 7.9 5.6 5.5 Thailand 3.2 3.2 4.0 Indonesia 2.3 - 2.2 China 3.1 2.7 2.1 " Include budgetary expenditures, extra-budgetary expenditures from educational surcharges, and SOE expenditures on education. Source: Ministry Of Education; State Statistical Yearbook (1995); ADB 1995; China: Social Sector Expenditures, World Bank, (1995) and World Bank (1994): East Asian Miracle report Overall spending on health has also been declining as a share of gross domestic product. While China's national health indicator compare favorably to other countries with similar per capita income levels, these indicators hide substantial regional variations. For example, infant mortality rates in the South Hinterland are nearly three times as high as those in coastal areas (see table A3.2). Access to adequate health care remains a problem for large segments of the population and affordability is a growing concern. 80 ANNEX 3 Inequalities in education and health care-both in terms of outcomes and expenditures- remain large (see Table A3.2). Provinces with lower per capita incomes and a higher proportion of minorities tend to have weaker social indicators. For instance, functional illiteracy rates in the Far West and South Hinterland are the lowest in China, twice those on the South Coast. The divergence in school attendance among the regions would predict persistence in these patterns. Table A3.2: Regional Education and Health Indicators, 1995 Minority Functional Basic 6-14 Year Population: Per capita Infant GDP Population Illiteracy Education olds in Natural Rate Health Mortality Per Capita Share (%) Rate (%) Expenditures School (%) of Increase Expenditure (per 1000) (per student) (%) (1994) yuan (1993) China 4,772 9.0 12.0 614 82.3 10.6 12.5 33.0 East Coast 8,848 0.4 11.5 1,335 89.3 5.0 16.6 17.6 South Coast 7,449 2.1 9.8 934 89.0 11.6 18.6 17.9 North Coast 5,557 5.0 11.4 456 87.6 5.3 10.8 14.9 North Hinterland 4,018 6.9 9.7 617 82.0 9.0 13.2 24.4 Far West 3,267 34.9 19.9 613 78.2 13.8 19.6 44.4 Central Core 3,467 2.9 12.1 764 82.7 8.6 7.7 43.8 South Hinterland 2,956 14.6 15.3 712 75.8 11.0 10.1 50.5 Source: World Bank (1995); 1996 State Statistical yearbook; 1995 Nian Quanguo 1 percent Renkou Chouyang Diaocha Zhuyao Shu Ju (Main Statistics from the 1995 one percent Population Census), SSB, 1995; 1995 Educational Statistics Yearbook of China; 1995 Health Yearbook of China. Disparities in government spending contribute to this outcome. These disparities are large and have grown with greater fiscal decentralization. The poorest regions with the weakest fiscal base have the lowest per capita expenditures, which are matched with the highest private fees. However, many of the poorer regions also have a large share of minorities, which adds to the cost of education. Guizhou spends less then 40 Yuan on basic education per capita while Shanghai's figure is over 200. Disparities among local budgetary health spending show a similar pattern. Whereas Anhui spends less than 7 Yuan per capita a year, Shanghai spends over 40. Inequalities in access and care have been exacerbated by changes in the structure of health care financing. The shift in health care finance towards insurance and insurance coverage tends to concentrate health care benefits on the urban population, especially government employees. Except for about 100 million of the highest earners, the rural population is uninsured. Over the reform period, health financing has shifted from the government budget and collective health schemes toward insurance and user fees. Whereas the government financed 25 percent of health expenditures in 1980, it only financed 13 percent in 1992-below levels in most Asian comparators. Rural collective funding declined even more sharply, from 16.7 percent of health expenditures in 1990 to 5.5 percent in 1992, thus leaving most of the rural population paying user fees. These fees now finance over a third of expenditures. The government can arrest a possible deterioration in China's social indicators, mitigate disparities in educational attainment and health status and promote growth by increasing the level of public expenditures allocated to basic education and health. To achieve the goal of universal access to good quality basic education, government expenditures on basic education would have to rise from the current 1.8 percent of gross domestic product to 2.9 percent of gross domestic product. Of the 1.1 percent rise in expenditures, about 0.7 percent is necessary to bring the 81 ANNEX 3 expenditures per student in basic education to acceptable levels.' The rest is necessary to increase the enrollment ratio from the current 82 percent to 100 percent. The package of public health and essential clinical services that is recommended in the World Development Report 1993 requires resources equivalent to 3.5 percent of gross domestic product for the whole population. A public health package alone-arguably a government responsibility-would require budgetary expenditures of 1.5 percent of GDP in 1995, more than three times the current government spending on health. Poverty Alleviation Poverty alleviation is a major priority for the Ninth Five-Year Plan period and China has raised expenditures on poverty alleviation commensurably. Public expenditures on poverty alleviation have increased rapidly in the past several years and are now comparatively higher than for education and health. But higher levels still are needed to meet the objectives of the National Seven-Year Plan for Poverty Reduction (8-7 Plan), to reach the rural poor residing outside the targeted 592 counties and to address the growing problems of the urban poor. The extension of the multi-pronged approach adopted in the Southwest Poverty Reduction Project to all the poor in China would necessitate annual expenditures of about Renminbi 18 billion or 0.33 percent of gross domestic product in 1995. This is only slightly more (0.26 percent of 1995 GDP) than the 15. billion planned for poverty spending annually for the rest of the Ninth Five-Year Plan period. Refinements in approach (such as targeting below the county level) could further improve the impact of the government's poverty alleviation efforts. Environment In addition to more effective regulatory control, in particular over TVEs, and appropriate pricing to dampen demand and reduce pollution, there is a need for allocating increased resources toward meeting the investment and operating costs of municipal services, including sewerage, wastewater treatment and solid waste management. China increased spending on pollution control, from 0.4 percent of GDP in 1980 to 0.67 percent in 1992, but this remains insufficient to deal with the nearly threefold increase in the output of heavy industry over the same period (international comparisons). Total suspended particulate (TSP) levels remain well above levels considered safe by the World Health Organization, sulfur dioxide concentrations exceed the lowest Chinese air quality standard. Few urban rivers reach the lowest acceptable Chinese water quality standard and groundwater quality continues to fall at an alarming rate. Petroleum discharges to surface waters are increasing after an initial decline. Only 4.5 percent of municipal wastewater flows receive treatment of any kind, while industrial pretreatment raises overall treatment rates to 17 percent. The Government's goal for the year 2000 is a modest 25 percent. Nearly 40 percent of urban China is unserved by sewers, with wastewater going directly into lakes and rivers, and according to current municipal investment plans, 30 percent of urban China will still remain unserved in the year 2000. We chose Zhejiang Province as representative of an "acceptable" expenditure level. Zhejiang is the non- city province with the highest expenditures after Guangdong. 82 ANNEX 3 Given the impact of untreated wastewater outside the producing municipalities- downstream effects and widespread aquifer pollution-central government intervention appears necessary to safeguard water quality in China. A wastewater investment program, which would provide treatment for any discharge that reduces receiving water quality to below irrigation use standards, would cost about 4.2 billion Yuan per year over the next ten years, requiring an increase of 15 percent in the annual urban infrastructure construction program. Improved solid waste disposal (through landfills) might cost about 5.2 billion yuan/year (8 in 95 prices) or 16 Yuan/capita/year-more than three times current charges for domestic waste. While the incremental cost of these interventions is a modest 0.3 percent of GDP (0.2 percent in 95 prices), cost recovery through increased user fees would result in a tripling of average effective water prices. Estimates of investments required to improve waste treatment and water supply in cities where there are severe shortages are in the order of RMB100 billion/year over the next 10 years or 1.7 percent of 1995 GDP. This compares with current urban water investments of only 0.35 percent of GDP. Substantial budgetary funds are expected to be necessary for inter-basin transfer projects (15 to 30 percent of project costs) and sewerage and industrial waste management (30 percent), yielding budgetary requirements of 0.3 percent of GDP. Infrastructure China needs greater public expenditures in infrastructure. This is not to deny that China has made tremendous strides in infrastructure investments in the last ten years. These amounted to 7.5 percent of GDP in 1994, up from 4.4 percent in 1985. Higher tariffs (e.g. power) and government-directed credit have been the principal sources of finance with only a small contribution from budgetary funds. This has resulted in underfunding of infrastructure projects that are not commercially viable but yield large economic benefits.2 Of particular concern, in this regard, is China's road infrastructure. Physical indicators show that China fares relatively poorly in comparison with other large countries in per capita coverage of the road network. It is therefore reasonable for China to aim for 8-9 percent of GDP and to channel increased budgetary funds of some 0.5-1.0 percent of GDP to infrastructure investments with public goods characteristics. Contingent Liabilities Public expenditure policies will have to address not only the underfunding of priority sectors discussed above, but also manage the implications of increased market orientation for the budget. Tasks previously performed by other sectors of the economy will increasingly be shifted to the budget, while in-kind benefits offered to civil servants will be monetized and contingent liabilities may increase. The implicit pension debt is estimated to equal around 50 percent of GDP and according to official estimates, about 20 percent of the state bank portfolio are nonperforming, which equals another 20 percent of GDP. In addition, with economic growth and rapidly rising wages, government services are likely to become relatively more expensive, necessitating a larger government share of the economy. In 1992, the budget financed 7.4 percent of infrastructure investments, a share marginally higher than the contribution of the budget to all state investments. China: Public Investment and Finance, World Bank, 1995. 83 ANNEX 3 Pension reforms appear to be gaining momentum. It is important to evaluate carefully the budgetary implications of various options. The government now appears to be favoring a multi-pillar approach, the first of which would have welfare and redistributive objectives and be funded on a pay as you go basis. Fully funded mandatory individual retirement accounts would consitute the second pillar and a voluntary scheme the third. A difficult issue that remains to be resolved concerns the transition from the current unfunded to a largely funded scheme in the future. Pension expenditures now account for some 2 percent of GDP and yield benefits that amount, on the average, to 70 percent of wages. The bulk of these payments (excluding civil servants) are now assumed by enterprises. Under the proposed scheme, the government would assume responsibility for the basic benefit, which the Bank proposes should equal 24 percent of wages, and transition costs. The first pillar would be financed out of a payroll tax equivalent to 6 percent of wages. The costs of transition in 1994 (1.3 percent of GDP) could be financed out of the surplus generated from the first pillar, a reduction of benefits and government borrowing or sale of SOE assets. Assuming an even split of the burden between government and SOEs would yield budgetary expenditures for 1994 of some 1.35 percent of GDP, 0.9 percent of which would be covered by the payroll tax. Unemployment benefits will rise with increasing open unemployment in urban areas. Assuming benefit levels equivalent to the public pillar of the pension scheme and 5 percent urban unemployment, budgetary outlays would equal 0.2 percent of GDP and could be financed by a payroll tax of 1-1.5 percent. Budgetary Expenditures Revisited. Adjusting the level of budgetary spending to reflect the need for additional expenditures in priority areas (as discussed above), and the implicit and extra-budgetary costs of all government activities yields budgetary expenditures in 1995 of some percent of GDP (Table A3.3). This is substantially higher than current budgetary expenditure levels and closer to international norms. Table A3.3: Proposed Expenditures: Cost of Shouldering Government Responsibility Expenditures Percent of GDP Actual Budgetary Expenditures,1995 Extra-budgetary Expenditures Additional Expenditures Proposed 5.35 Education (1.1) Health (1.5) Poverty Alleviation (.05) Environment (0.5) Infrastructure (0.7) Social Insurance (1.5) Adjusted Budgetary Expenditures, 1994 84 ANNEX 4 ANNEx 4: CHINA'S FISCAL PROSPECTS IN THE MEDIUM TERM' Summary The Staff's medium-term economic projections assume that the macro-economic performance achieved recently can be sustained in the medium-term. Combined with an assumption of unchanged tax policies, the revenue-to-GDP ratio can be expected to increase only modestly (probably less than one percentage point) in the years to 2010. The revenue ratio is projected to increase by about 3/4 percentage point to the year 2010, mainly as a result of a continued increase in household consumption and hence in consumption taxes and a buoyant personal income tax. Significant off-budget expenditures in China are financed through the social funds and the EBFs. According to some estimates, the related revenues and expenditures constitute close to 7 percent of GDP. Against this background, a medium-term fiscal framework must necessarily build upon the consolidated operations of the state budget and all off-budget funds. The goals and objectives for the medium-term for key sectors of the economy, as stated by the Chinese authorities, imply that the budget will probably need to assume significant new expenditure burdens if the present momentum of economic development is to be sustained. Sizable expenditure demands are expected, mainly in three broad areas. Increases in physical infrastructure investment demands are foreseen for housing and highway construction, agricultural development and power generation. Similarly, significant social infrastructure demands will result from the establishment of a modern social safety net, including pension and unemployment funds. Large additional resources will also have to be devoted to financial infrastructure investment, primarily to clean up bad debt in the banking sector with a view to creating a sound financial basis for a market-based economy. The resulting overall expenditure demands can only be projected with considerable uncertainty and will almost entirely depend on specific political decisions. Hence, the numbers used in the medium-term projection within the different areas should be considered merely as technical working hypotheses at this stage. To the extent that the prospective expenditure demands, whether deriving from the moveon- budget of extra-budgetary activities or from new expenditure demands, will affect the fiscal balance, the authorities must seek to assess the possible magnitude of the resulting financing needs. Following the need to maintain the objective of a zero fiscal deficit in the medium-term, the prospective expenditure demands will have to be met by new revenue measures. The numbers used in this annex differ slightly from those of the main text and of Annex 3, due to different assumptions and definitions. 85 ANNEX 4 The Macroeconomic Framework The basis of the fiscal projection to the year 2010 presented below is a macro-economic framework which basically assumes that the impressive macro-performance achieved in the most recent years can be sustained in the medium term, characterized by a high level of non-inflationary growth of about 9 percent from 1998 and onwards (see Table A4.1 for details). On the demand side, private household consumption is assumed to continue its increase relative to GDP, whereas public consumption and fixed investments are assumed to stabilize as a percentage of GDP. The level of openness of the Chinese economy is assumed to continue to increase, albeit only modestly, as reflected in growing import-and export ratios, from about 18-19 percent in 1997 to about 21 percent in 2010. Key to the projection of fiscal revenues is the expected developments in the functional distribution of income. In this regard, the projection assumes that aggregate wage incomes, concurrently with the maturing of the Chinese economy, will constitute a growing share of total factor incomes, increasing from the present (and very low) level of about 15 percent to about 20 percent over the next 15 years. A Fiscal Benchmark Projection Generally, in the short run, there is no simple and stable relationship between tax bases and macro-economic (SNA) aggregates. Over the medium- and long-term, however, it is reasonable to expect that systematic changes in macro-aggregates, under an assumption of unchanged policies, will be reflected in corresponding changes in tax bases, and hence in tax yields from consumption, income and trade taxes. The projection presented here (see Table A4.2) assumes such a relationship. Concerning income taxes, the personal income tax is assumed to be positively affected by the relative increase in wages, as well as by the nominally fixed tax threshold, which have significantly contributed to the high buoyancy of this tax in recent years (with yields increasing by almost 50 percent in 1996). With revenues at 0.3 percent of GDP, the PIT at present contributes very little to overall revenues and its high buoyancy cannot alter the fact that overall taxes related to income and profits (including the enterprise income tax) will continue to contribute less than 2 percent of GDP in the projection period. Regarding consumption taxes (VAT + excises), the assumed increase in private household consumption will tend to marginally lift the yield, and over the whole of the projection period the projection foresees an increase in these taxes of only about 0.5 percent of GDP. A similar development is projected fortrade taxes, resulting from the assumed increase in the import ratio. Over the period of time discussed here, expenditure developments can broadly be considered the cumulative outcome of explicit policy choices rather than a reflection of underlying structural trends or economic cycles. This highlights the importance of the assumptions made concerning the nature of "unchanged policies". In the present projection, subsidies (to SOEs and price subsidies) are assumed to stabilize at the historically low level of about one percent of GDP from 1997 and onwards. Owing to the strong expansion of bond financing (bond issuance is expected to increase by about 25 percent from 1996 to 1997), interest expenses in the budget have increased and is expected to continue to do so albeit from a very low level. Because of the lowering of the state budget deficit 86 ANNEX 4 Table A4.1: Macro Framework 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Nominal GDP 3463.44 4662.23 5826.05 6780 7869.55 9092.47 10505.44 12137.99 14024.23 16203.6 18721.64 21630.99 24992.46 28876.79 33363.68 38548.41 44538.84 51460.19 Growth in percent 30.02 34.61 24.96 16.37 16.07 15.54 15.54 15.54 15.54 15.54 15.54 15.54 15.54 15.54 15.54 15.54 15.54 15.54 Real GDP (1990 basis) 2627.72 2598.81 3269.49 3586.63 3927.36 4280.82 4666.1 5086.04 5543.79 6042.73 6486.58 7197.37 7825.51 8529.81 9297.49 10134.26 11046.35 12040.52 Growth in percent 13.5 12.6 10.5 9.7 9.5 9 9 9 9 9 9 9 9 9 9 9 9 9 Demand components: Private consumption 1631.5 2199.07 2813.24 3237.22 3807.64 4443.11 5152.79 5999.12 7003.14 8192.05 9521.26 11044.13 12810.38 14858.87 17234.67 19999.04 23185.57 26891.54 Public consumption 424.2 538.50 635 721 826.3 972.89 1155.6 1359.45 1584.74 1831.01 2115.55 2444.31 2824.15 3263.03 3770.1 4355.98 5032.9 5815.01 Fixed investment 1298.0 1685.6 2000.1 2406.9 2770.08 3182.37 3624.38 4187.61 4838.36 5590.24 6458.97 7462.69 8622.4 9962.32 11510.47 13299.2 15365.9 17753.76 Private Public Inventory investment 201.8 173.6 277.91 332.22 409.22 454.62 525.27 546.21 560.97 567.13 608.46 670.57 737.28 808.54 884.15 963.72 1046.67 1132.14 Exports (goods and serv) 693.87 1026.18 1229.66 1281.78 1509.76 1727.78 1987.5 2282.6 2617.69 3000.89 3523.39 4135.82 4853.51 5694.38 6679.37 7833 9183.86 10765.42 Imports (goods and serv) 785.97 960.74 1129.8 1199.12 1453.46 1688.3 1940.1 2237 2580.66 2977.71 3505.97 4126.51 4855.25 5710.82 6715.06 7893.5 9276.03 10897.64 Functional income components: Wage bill 491.6 665.6 850.98 1012.69 1201.4 1418.1 1673.14 1973.2 2326.12 2741.07 3228.82 3801.96 4475.26 5266.01 6194.44 7284.27 8563.23 10063.78 Profits, gross 2971.84 3996.63 4975.07 5767.31 6668.15 7674.37 8882.3 10164.79 11698.11 13462.53 15462.83 17829.04 20517.2 23610.29 27169.24 31264.13 35975.61 41396.42 Check (GDP-C-I-(X- 0.04 0.02 -0.06 0 0.01 -0.01 0 0 -0.01 -0.01 -0.01 -0.01 -0.02 -0.02 -0.02 -0.02 -0.03 -0.03 M)=0 GDP ratios: 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Private consumption 0.47 0.47 0.48 0.48 0.48 0.49 0.49 0.49 0.5 0.51 0.51 0.51 0.51 0.51 0.52 0.52 0.52 0.52 Public consumption 0.12 0.12 0.11 0.11 0.1 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 Fixed investment 0.37 0.36 0.34 0.36 0.35 0.35 0.35 0.35 0.35 0.34 0.34 0.34 0.34 0.34 0.34 0.34 0.34 0.34 Private Public Inventory investment 0.06 0.04 0.05 0.05 0.05 0.05 0.05 0.05 0.04 0.4 0.03 0.03 0.03 0.03 0.03 0.03 0.02 0.02 Exports (goods and serv) 0.2 0.22 0.21 0.19 0.19 0.19 0.19 0.19 0.19 0.19 0.19 0.19 0.19 0.2 0.2 0.2 0.21 0.21 Imports (goods and serv) 0.23 0.21 0.19 0.18 0.18 0.19 0.18 0.18 0.18 0.18 0.19 0.19 0.19 0.2 0.2 0.2 0.21 0.21 Wage bill 0.14 0.14 0.15 0.15 0.15 0.16 0.16 0.16 0.17 0.17 0.17 0.18 0.18 0.18 0.19 0.19 0.19 0.2 Profits, gross 0.86 0.86 0.85 0.85 0.85 0.84 0.84 0.84 0.83 0.83 0.83 0.82 0.82 0.82 0.81 0.81 0.81 0.8 87 ANNEX 4 Table A4.2: Benchmark Projection - The State Budget 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Revenues 475.9 558.4 654 741.2 871.15 1015.74 1177.79 1369.27 1593.36 1856.62 2156.44 2501.58 2901.97 3364.49 3900.67 4522.26 5242.83 6078.15 Tax revenues 447.5 524 605.5 684 804.76 939.03 1089.16 1266.87 1475.05 1719.91 1998.49 2319.08 2691.12 3120.87 3619.2 4197.04 4867.08 5644 Income & profit taxes 80.5 94 110.9 123.55 145.54 170.9 199.18 232.21 269.46 312.72 362.97 421.34 489.15 565.95 654.8 757.6 876.54 1014.15 Enterprise income taxes ... ... ... 104.55 120.69 138.91 159.86 183.98 211.74 243.67 280.42 322.71 371.36 427.35 491.76 565.88 651.16 749.28 Personal income taxes ... ... ... 19 24.84 31.99 39.32 48.23 57.72 69.05 82.55 98.64 117.79 138.6 163.04 191.72 255.38 264.88 Consumption taxes 287.8 346.5 399 461.75 543.12 633.76 734.99 855.71 998.92 1168.5 1358.1 1575.32 1827.26 2119.45 2548.33 2851.35 3307.16 3835.78 VAT 108.1 230.8 258.6 298.29 350.85 409.41 474.8 552.78 645.3 754.85 877.33 1017.65 1180.4 1369.15 1588.07 1841.96 2136.41 2477.89 Other 179.7 115.7 140.4 163.46 192.27 224.35 260.19 302.92 353.62 413.66 480.77 557.67 646.86 750.3 870.26 1009.39 1170.75 1357.88 Customs duties 25.6 27.3 28.7 30 36.36 42.24 48.54 55.97 64.56 74.5 87.71 103.24 121.47 142.88 168 197.48 232.07 272.64 Other taxes 53.6 56.2 66.9 68.7 79.74 92.13 106.45 122.99 142.1 164.19 189.7 219.18 253.24 292.6 338.07 390.6 451.3 521.43 Non-tax revenues 28.4 34.4 48.5 57.2 66.39 76.71 88.63 102.4 118.32 136.7 157.95 182.49 210.85 243.62 281.48 325.22 375.76 434.15 Expenditures 545.9 632.6 751.8 867.96 1000.25 1164.78 1345.79 1554.93 1782.54 2051.45 2360.88 2716.95 3126.66 3598.11 4140.58 4764.75 5482.92 6309.24 Current 424.3 538.5 624 726.36 842.86 982.93 1135.68 1312.17 1502.06 1727.37 1986.45 2284.33 2626.81 3020.58 3473.3 3993.78 4592.15 5280.04 Subsidies 71 68 71.7 79.32 78.7 90.92 105.05 121.38 140.24 162.04 187.22 216.31 249.92 288.76 333.64 385.48 445.39 514.6 Interestexpenses 9.7 16.7 35.6 49.7 70.83 90.92 105.05 121.68 126.22 137.73 149.77 162.23 174.95 187.7 200.18 212.02 222.69 231.57 Other 343.6 453.8 516.7 597.35 693.34 801.09 925.57 1069.41 1235.59 1427.61 1649.46 1905.78 2201.94 2544.13 2939.48 3396.28 3924.06 4533.86 Capital 121.6 94.1 127.8 141.6 157.39 181.85 210.11 242.76 280.48 324.07 374.43 432.62 499.85 577.53 667.27 770.97 890.78 1029.2 Balance -70 -74.2 -97.8 -126.76 -129.1 -149.05 -168 -185.65 -189.18 -194.83 -204.44 -215.37 -224.69 -233.62 -239.9 -242.49 -240.09 -231.09 Ratios to GDP: Revenues 13.74 11.98 11.23 10.93 11.07 11.17 11.21 11.28 11.36 11.46 11.52 11.56 11.61 11.65 11.69 11.73 11.76 11.81 Tax revenues 12.92 11.24 10.39 10.09 10.23 10.33 10.37 10.44 10.52 10.61 10.67 10.72 10.77 10.81 10.85 10.89 10.93 10.97 Income taxes 2.32 2.02 1.9 1.82 1.85 1.88 1.9 1.91 1.92 1.92 1.93 1.94 1.95 1.96 1.96 1.96 1.97 1.97 Enterprise income taxes 0 0 0 1.54 1.53 1.53 1.52 1.52 1.51 1.5 1.5 1.49 1.49 1.48 1.47 1.47 1.46 1.46 Personal income taxes 0 0 0 0.28 0.32 0.35 0.37 0.4 0.41 0.43 0.44 0.46 0.47 0.48 0.49 0.5 0.51 0.51 Consumption taxes 8.31 7.43 6.85 6.81 6.9 6.97 7 7.05 7.12 7.21 7.25 7.28 7.31 7.34 7.37 7.4 7.43 7.45 VAT 3.12 4.95 4.44 4.4 4.46 4.5 4.52 4.55 4.6 4.66 4.69 4.7 4.72 4.74 4.76 4.78 4.8 4.82 Other 5.19 2.48 2.41 2.41 2.44 2.47 2.48 2.5 2.52 2.55 2.57 2.58 2.59 2.6 2.61 2.62 2.63 2.64 Customs duties 0.74 0.59 0.49 0.44 0.46 0.46 0.46 0.46 0.46 0.46 0.47 0.48 0.49 0.49 0.5 0.51 0.52 0.53 Other taxes 1.55 1.21 1.15 1.01 1.01 1.01 1.01 1.01 1.01 1.01 1.01 1.01 1.01 1.01 1.01 1.01 1.01 1.01 Non-tax revenues 0.82 0.74 0.83 0.84 0.84 0.84 0.84 0.84 0.84 0.84 0.84 0.84 0.84 0.84 0.84 0.84 0.84 0.84 Expenditures 15.76 13.57 12.9 12.8 12.71 12.81 12.81 12.81 12.71 12.66 12.61 12.56 12.51 12.46 12.41 12.36 12.31 21.26 Current 12.25 11.55 10.71 10.71 10.71 10.81 10.81 10.81 10.71 10.66 10.61 10.56 10.51 10.46 10.41 10.36 10.31 10.26 Subsidies 2.05 1.46 1.23 1.17 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Interest expenses 0.28 0.36 0.61 0.73 0.9 1 1 1 0.9 0.85 0.8 0.75 0.7 0.65 0.6 0.55 0.5 0.5 Other 9.92 9.73 8.87 8.81 8.81 8.81 8.81 8.81 8.81 8.81 8.81 8.81 8.81 8.81 8.81 8.81 8.81 8.81 Capital 3.51 2.02 2.19 2.09 2 2 2 2 2 2 2 2 2 2 2 2 2 2 Balance -2.02 -1.59 -1.69 -1.87 -1.64 -1.64 -1.6 -1.53 -1.35 -1.2 -1.09 -1 -0.9 -0.81 -0.72 -0.63 -0.54 -0.45 88 ANNEX 4 and the trend towards declining interest levels, interest expenses are assumed to reach a maximum of about one percent of GDP at the end of the century and subsequently to decline at the end of the projection period. Other current expenditures are assumed to stabilize at close to 9 percent of GDP and capital expenditures are assumed to further decline and then stabilize at a historically very modest level of two percent of GDP. Under these assumptions, the revenue ratio will increase over the projection period by a modest 0.75 percent of GDP, while expenditures can be expected to decline marginally by about 0.5 percent of GDP, leaving a deficit of about 0.5 percent of GDP in 2010. In other words, assuming unchanged policies, there are factors which will tend to improve the fiscal balance, an improvement estimated at a shade over one percent of GDP at the end of the projection period compared to the actual present level. Sizable fiscal operations in China take place off-budget. This holds for the operations of the five social funds and the EBFs at both the central and the local levels. According to GFS principles, the majority of these operations should clearly be included in the recording of the operations of general government. Since these funds-according to some estimates-have reached an overall level close to 7 percent of GDP, and since they conduct key functions in the areas of social policies and capital spending, serious distortions would follow if they were excluded from the context of a fiscal projection. Consequently, Table A4.3 presents a projection including these funds, i.e., for the general government as a whole. Since the fiscal recording in China is particularly deficient as regards to the nature and size of the operations of these segments of general government, the basis for making a medium-term projection is-at best-incomplete. The projection presented here is based on a "mechanistic" assumption of unchanged activities relative to GDP. The official policy is to reduce the scope of the extra-budgetary activities, by either curtailing or moving them on-budget. Thus, following a decision by the State Council, in 1997, 13 largeEBFs (or fees) at the central level will be moved on-budget, with a corresponding increase of revenue and expenditure in the state budget of about 1.5 percent of GDP. Also, following a major investigation of the nature and size of the EBFs conducted during 1996, the authorities plan to impose strict limits on the operations of the local EBFs in the future. Consequently, the balance between the official budget and the EBFs will presumably change dramatically over the medium-term, with a marked decline in the size of extra-budgetary operations and presumably, a broadly corresponding increase in the volume of the recorded state budget. In the projection presented here, it is basically assumed that the present level of the fiscal activities conducted by these funds will be carried on, whether recorded on/or off-budget. This assumption may underestimate the potential efficiency gains and explicit expenditure cuts that may follow from the increased scrutiny of the funds' operations, presumably resulting from their move on-budget. Under the assumptions made, the revenues and expenditures of the consolidated general government in the benchmark scenario will develop as illustrated in Table A4.3, with general revenues increasing from a current level of about 18 percent of GDP to 18.75 percent at the end of the projection period, and with expenditures declining by about 0.5 percent of GDP to 19 percent in 89 ANNEX 4 Table A4.3: Benchmark Projection - Consolidated General Government 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 State budget: Revenue 475.9 558.4 654 741.2 871.15 1015.74 1177.79 1369.27 1593.36 1856.62 2156.44 2501.58 2901.97 3364.49 3900.67 4522.26 5242.83 6078.15 Expenditure 545.9 632.6 751.8 867.96 1000.25 1164.78 1345.79 155493 178254 2051 45 2360.88 2716.5 3126.66 3598.11 414058 4764.75 548292 63024 Balance -70 -74.2 -97 8 -126 76 -129.1 -149.05 -168 -18565 -189.18 -19483 -20444 -215.37 -22469 -233.62 -2399 -242.49 -24009 -231.09 Social funds: 1/ Revenue 63 82 702 73.71 85 78 99 56 115.04 132.91 153.57 17743 205 236.86 273.67 316.2 365.34 422.11 417 71 5635 651 06 Expenditure 57.91 63.7 66.89 7784 90.35 104.38 120.61 13935 161 186.02 21493 248.33 286.92 331.51 383.03 44255 511.32 59078 Balance 5.91 65 6.82 794 9.22 1065 12.31 1422 16.43 18.98 21.93 25.34 2928 33.83 3903 45.16 52.18 6028 Extra-budgetary funds: Revenue 143.25 186.25 240 3843 446.06 515.37 59546 688 794.91 918.44 1061.17 1226.08 1416.61 1636.75 1891.1 2184.98 2542.52 291684 Expenditure 131.43 171.09 240 3843 446.06 515.37 59546 688 794.91 918.44 1061.17 1226.08 141661 1636.75 1891.1 2184.98 2542.52 2916.84 Budgetary 11.82 15.16 0 0 0 0 0 0 0 0 Consolidated general government: Revenue 682.97 814.85 967.71 1211.28 141677 1646 15 1906 16 2210.84 2565.71 2980.06 3454.47 4001.32 4634.78 5366.57 6213.88 7194.4 8330.85 964605 Expenditure 735.24 867.39 1058.69 1330.1 1536.66 178454 2061.86 2382.28 2738.46 3155.91 3636.98 4191.35 4830.19 5366.57 6414.7 7392.37 8518.77 9816.86 Balance -52.27 -52.54 -90.98 -118.81 -119.88 -138.4 -1557 -171 43 -172.75 -175.85 -182.51 -190.03 -19542 -199.8 -200.82 -197.33 -187.92 -1708 Ratios to GDP State budget: Revenue 13.74 11.98 11.23 10.93 11.07 11.17 11.21 11.28 11.36 11.46 11.52 11.56 11.61 11.65 11.69 11.73 11.77 11.81 Expenditure 15.76 13.57 12.9 12.8 1271 12.81 12.81 12.81 12.71 12.66 12.61 12.56 12.51 12.46 12.41 12.36 12.31 12.26 Balance -2.02 -1.59 -1.68 -1.87 -1 64 -1.64 -1.6 -1 53 -1.35 -1.2 -1.09 -1 -0 9 -0.81 -0.72 -0.63 -0.54 -0.45 Social budget: Revenue 1 84 Expenditure 1.67 Balance 0 17 Extra-budgetary funds: Revenue 4.14 3.99 4.12 5.67 567 5.67 5.67 5 67 5.67 567 - 567 567 5.67 5.67 5.67 5 67 5.67 5.67 Expenditure 3.79 3.67 4.12 5.67 5.67 5.67 567 567 5,67 5.67 5.67 5.67 5.67 5.67 567 5.67 5.67 5.67 Budgetary 0.34 0.33 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Consolidated general government: Revenue 19.72 17.48 16.61 17.87 18 18.1 18.14 18.21 18.29 18.39 18.45 18.5 18.54 18.58 18.62 18.66 18.7 18.74 Expenditure 21.23 18.6 18.17 19.62 19.53 19.63 19.63 19.63 19.53 19.48 19.43 19.38 19.33 19.28 19.23 19.18 19.13 19.08 Balance -1.51 -1.13 -1.56 -1.75 -1.52 -1.52 -1.48 -1 41 -1.23 -1.09 -0.97 -0.88 -0.78 -0.69 -0.6 -0.51 -0.42 -0.33 If Staff estimates Includes the pension funds, unemployment funds, medical fund, injury fund, and maternity fund. 90 ANNEX 4 the year 2010. This would imply a general government deficit of less than 0.5 percent of GDP, a decline of more than one percent of GDP from the present level. Prospective Expenditure Demands Consistent with officially stated policy objectives for the near- and medium-term, the budget will have to carry significant new expenditure programs. Little concrete information is available on the specificities of the nature, size or sequencing of these programs, and their fiscal implications can only be estimated as very rough orders of magnitude (see Table A4.4). Physical Infrastructure Significant infrastructure investment projects are planned in the areas of housing, roads and communications, agricultural improvements, energy and the environment. According to government estimates, the new highway construction program will cost approximately $150 billion over the next 15 years, of which approximately 20 percent is expected to be funded by the state budget, while the rest will be the responsibility of the provinces. It is uncertain to which extent this program will be implemented in its entirety, and with what burdens for the budget. Even larger construction projects are planned in the area of energy and power generation, to some extent with active government co-participation (through foreign loans guarantees or with the state budget directly assuming the debt burden), but in principle they are assumed to be self-financing over their economic lifetime. Because of the uncertainty relating to the possible budgetary burdens, the energy area has been held outside the projection presented here.6 In the area of environmental improvement, significant prospective expenditure needs will have to be addressed according to official statements. The World Bank has estimated these needs to be about 0.2 percent of GDP over currently budgeted level. Concerning the totality of required infrastructure investments, in part assessed on the basis of the experiences of other developing countries, the World Bank has estimated that in order to sustain the present growth momentum and to avoid serious infrastructural bottlenecks, China would need to raise the investment ratio to about 7-8 percent of GDP. Social Infrastructure This heading covers a number of key expenditure areas where prospective demands will be significant if the momentum of economic growth is to be maintained with a minimum of social disruption. The associated future costs will depend on the policy decisions by the government concerning the design of a new social safety net system, the speed and strength of SOE reform measures, etc. In the area of pension reform, the World Bank has estimated that a move to a multi-pillar, partly funded pension system may entail transitory costs (present value of contingent liabilities during the phasing-out of the present pay-as-you-go system) of up to 50 percent of GDP, in addition to the costs of the new system. In the 1996 CEM, the Bank estimated the financing gap following the transition to the new system to slightly above one percent of GDP on an annual basis, to be covered in part by increased sales of state assets. 91 ANNEX 4 Table A4.4: Prospective Additional Budgetary Expenditures - General Government 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Physical infrastructure ... ... ... ... ... 45.46 63.06 84.97 112.19 145.83 187.22 237.94 299.91 375.39 467.09 578.23 668.08 771.9 Housing construction . ... ... .. ... Roads and other infrastructure ... ... ... ... ... 20 23.11 26.7 30.85 35.65 41.19 47.59 54.98 63.53 73.4 84.81 97.99 113.21 Agriculture ... ... ... . E nergy ... ... ... ... ... ... ... .. .... .. .. Environment ... ... ... ... ... 18.18 21.01 24.28 28.05 32.41 37.44 43.26 49.98 57.75 66.73 77.1 89.08 102.92 Social infrastructure ... ,.. ... ... ... 99.92 193.71 316.69 457.19 625.46 778.82 964.74 1139.66 1345.64 1588.11 1834.9 2120.05 2449.51 Pension system ... ... ... ... ... 27.28 63.03 109.24 168.29 243.05 280.82 324.46 374.89 433.14 500.46 578.23 668.08 771.9 Extended coverage ... ... ... ... ... 18.18 42.02 72.83 112.19 162.04 187.22 216.31 249.92 288.76 333.64 385.48 445.39 514.6 Transition costs ... ... ... ... ... 9.09 21.01 36.41 56.1 81.02 93.61 108.15 124.96 144.38 166.82 192.74 222.69 257.3 Unemployment, ... ... ... ... ... 4.55 10.51 18.21 28.05 32.41 37.44 43.26 49.98 57.75 66.73 77.1 89.08 102.92 reemployment Other social safety net ... ... ... ... Poverty alleviation ... ... ... ... 29.15 40.82 57,14 80 92.56 106.94 123.56 142.76 164.95 190.58 220.2 254.42 293.96 339.64 SOE reform ... ... Health ... ... ... ... ... 18.18 42.02 72.83 112.19 162.04 224.66 302.83 349.89 404.27 467.09 539.68 623.54 720.44 Education ... ... ... ... ... 9.09 21.01 36.41 56.1 81.02 112.33 151.42 199.94 259.89 333.64 385.48 445.39 514.6 Financial infrastructure ... ... .., ... ... 9.09 21.01 36.41 56.1 81.02 93.61 108.15 124.96 144.38 166.82 192.74 222.69 257.3 Bank recapitalization ... ... ... ... ... 9.09 21.01 36.41 56.1 81.02 93.61 108.155 124.96 144.38 166.82 192.74 222.69 257.3 Total additional ... ... ... ... ... 154.47 277.76 438.07 625.48 852.31 1059.65 1310.84 1564.53 1865.41 2222.02 2605.87 3010.83 3478.71 expenditures Percent ofGDP: 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Physical infrastructure ... ... ... ... ... 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.5 1.5 Housing construction ... ... ... ... ... Roads and other infrastructure ... ... ... ... ... 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22 Agriculture ... ... ... ... ... Energy.. .. .. .. .... .. .. .. .. ............ Environment ... ... ... ... ... 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 Social infrastructure ... ... ... ... ... 1.1 1.84 2.61 3.26 3.86 4.16 4.46 4.56 4.66 4.76 4.76 4.76 4.76 Pension system ... ... ... ... ... 0.3 0.6 0.9 1.2 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 Extended coverage ... ... ... ... ... 0.2 0.4 0.6 0.8 1 1 1 1 1 1 1 1 1 Transition costs ... ... ... ... ... 0.1 0.2 0.3 0.4 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 Unemployment, ... ... ... ... ... 0.05 0.1 0.15 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 reemployment Other social safety net ... ... ..... Poverty alleviation ... ... ... ... ... 0.45 0.54 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 SOE reform ... ... ... ... ... Health ... ... ... ... ... 0.2 0.4 0.6 0.8 1 1.2 1.4 1.4 1.4 1.4 1.4 1.4 1.4 Education ... ... ... ... ... 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1 1 1 Financial infrastructure ... ... ... ... ... 0.1 0.2 0.3 0.4 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 Bank recapitalization ... ... ... ... ... 0.1 0.2 0.3 0.4 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 Total additional ... ... ... ... ... 1.7 2.64 3.61 4.46 5.26 5.66 6.06 6.26 6.46 6.66 6.76 6.76 6.76 expenditures 92 ANNEX 4 Table A4.5: Medium-Term Financing Gap - Consolidated General Government 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Billions of Yuan Revenues, benchmark proj. 682.97 814.85 967.71 1211.28 1416.77 1646.15 1906.16 2210,84 2565.71 2980.06 3454.47 4001.32 4634.78 5366.57 6213.88 7194.94 8330.85 9646.05 Expenditure, total 735.24 867.39 1058.69 1330.1 1536.66 1939.02 2339.62 2820.35 3363.94 4008.22 4696.62 5502.19 6394.72 7431.78 8636.72 9988.15 11529.6 13295.6 Benchmark proj. 735.24 867.39 1058.69 1330.1 1536.66 1784.54 2061.86 2382.28 2738.46 3155.91 3636.98 4191.35 4830.19 5566.37 6414.7 7392.27 8518.77 9816.86 Additional exp. ... ... ... ... ... 154.47 277.76 438.07 625.48 852.31 1059.65 1310.84 1564.53 1865.41 2222.02 2605.87 3010.83 3478.71 Balance, financing gap -52.27 -52.54 -90.98 -118.81 -119.88 -292.87 -433.45 -609.5 -798.23 -1028.16 -1242.16 -1500.87 -1759.94 -2065.2 -2422.84 -2803.21 -3198.74 -3649.51 Percent of GDP Revenues, benchmark proj. 19.72 17,48 16.61 17.87 18 18.1 18.14 18.21 18.29 18.39 18.45 18.5 18.54 18.58 18.62 18.66 18.7 18.74 Expenditure, total 21.23 18.6 18.17 19.62 19.53 21.33 22.27 23.24 23.99 24.74 25.09 25.44 25.59 25.74 25.89 25.94 25.84 25.84 Benchmark proj. 21.23 18.6 18.17 19.62 19.53 19.63 19.63 19.63 19.53 19.48 19.43 19.38 19.33 19.28 19.23 19.18 19.13 19.18 Additional exp. 0 0 0 0 0 1.7 2.64 3.61 4.46 5.26 5.66 6.06 6.26 6.46 6.66 6.76 6.76 6.76 Balance, financing gap -1.51 -1.13 -1.56 -1.75 -1.52 -3.22 -4.13 -5.02 -5.69 -6.35 -6.63 -6.94 -7.04 -7.15 -7.26 -7.27 -7.18 -7.09 93 ANNEX 4 Following reform of the SOE sector, substantial expenditure demands can be expected with regard to unemployment benefits and re-employment activities. Similarly, SOE reform measures per se may require budgetary resources, which are extremely difficult to project with a reasonable level of precision. The fight against poverty will, according to official policies, assume a very high priority in the medium-term. Related expenditures are projected by the Government to increase by about 40 percent per year, to reach about 100 billion Yuan in the year 2000. This estimate is used in the projection. Partly related to the problem of alleviating poverty, significant increases are planned for educational and health expenditures. To achieve universal access to good quality education and health services, the World Bank has estimated an additional annual expenditure requirement of about 2.3 percent of GDP. Financial Infrastructure Improving the soundness of the banking system will almost invariably lead to a demand for budgetary resources to clean up the portfolios of both the specialized banks and the policy banks. According to some reports, the authorities have already initiated this operation by increasing the reserves for bad debts linked to bankruptcies to 30 billion Yuan in 1997 (0.4 percent of GDP) from 20 billion Yuan last year, presumably in the context of the central budget. The present stock of bad debts in the banking system is considerable. The budgetary costs associated with such an operation will totally depend on how and how quickly it is executed. The prospective expenditure demands associated with these programs are extremely difficult to project with a reasonable level of certainty. Consequently, the numbers presented in Table A4.4 should merely be considered as expressing technical working hypotheses. However, whatever the specific composition of the demands may be, their overall level is significant. Thus, a required lift of expenditures on the order of 5-8 percent of GDP does not seem implausible, resulting in financing of a closely similar magnitude at the end of the projection period, as shown in Table A4.5. Without adequate financing, such a gap would seriously threaten economic stabilization and revive inflationary pressures. Financing Medium-Term Expenditure Demands A financing gap of this size can only be met through fundamental changes in financing policies and tax structures, mainly through a combination of three sources: borrowing, sales/leasing of state assets ("privatization") or increased taxation. While the actual choice of balance between the three sources is a matter of political priority, it is assumed here that the tax-financing component will be dominant. While difficult to factor in, it is assumed here that improvements in tax administration will continue along the lines planned by the authorities, with the establishment of a modem and computerized administration during the projection period. Under that assumption, some combination of the following tax policy measures would be necessary to finance the required level of revenues. * Scaling-back of tax concessions under the corporate income tax (CIT): gradually increasing the CIT rate in SEZs and other concession areas; unification of banking taxation of domestic and foreign banks; tightening tax holiday provisions; withholding taxes on dividend remittances (if tax treaty network allows it); repeal deductibility of financing of EBFs (if still present); repeal concessions on CIT liabilities to SOEs 94 ANNEX 4 (subsidies); introduction of standard accounting rules for calculating enterprise profits to avoid distortive taxation * Broadening the base of the personal income tax: implement a phased reduction in the basic threshold; broadening of tax base by including self-employed and capital income, particularly interest income, dividend and capital gains * Broadening the scope of payroll taxation to finance the expanded coverage of social safety net funds * VAT: extending base to services; repeal exemptions and rebates; repeal lower rate * "Normalize" the VAT (zerorating of exports, input credits on capital) * Excises: increase excise on gasoline; increase excise on cars and other luxury items * Other "green" excises * Strengthening of land, property, and inheritance taxation * Clarify taxation relations between levels of government: unify treatment of enterprises with different ownership forms; repeal exemptions granted by local governments; clarify tax sharing and transfer mechanisms These measures will significantly improve the revenue buoyancy of the Chinese tax system and reduce present distortions. 95 ANNEX 5 ANNEX 5: FURTHER STRENGTHENING THE MANAGEMENT OF EXTRA-BUDGET FUNDS -ANY LOCALITY, DEPARTMENT OR UNIT SHOULD NOT HAVE THEIR FISCAL REVENUE CONCEALED AND UNREPORTED People's Daily. 6 August, 1996, Page 1 Xinhua News Agency, 5 August 1996: The State Council made a decision recently on strengthening the management of extra-budget funds. According to the information, since the reform and opening up policy was adopted, the extra-budget funds in our country increased rapidly, which played an active role in the economic construction and social welfare development. However, in recent years, some localities, at their own will, have transferred fiscal- budget funds to extra-budget funds through various kinds of illegal means, which was against the Budget Law of the People 's Republic of China and the relevant regulations issued by the State Council. Some departments and units established their fund or fee programs at their own wish, which caused the loss of the state fiscal revenue and the continuous expansion of the extra- budget funds. At the same time, with the imperfect management system, the utilization of the extra-budget funds is not under the fiscal management and the supervision of the various levels of people's congresses. The problems of disorderly spending and over spending became severe. Those problems have not only caused the scattering of the state fiscal funds and the mass of the government public distribution order, but also expanded the fixed assets investment and the consumption funds, which contributed to the unhealthy tendencies and the corruption. Therefore, the State Council asked, in the decision, that all localities, departments and units should strictly follow the Budget Law of the People's Republic of China. It is forbidden to transfer budget funds to the extra-budget. All levels of people's government should strengthen the management over the fiscal-budget funds and extra-budget funds, and perfect the supervision and checking system on fiscal funds. Any locality, department or unit should not have their fiscal revenue concealed and unreported and transfer the fiscal-budget funds to extra-budget funds. Without approval from the fiscal departments, all departments and units should not change their fiscal appropriations to compensatory utilization purposes at their will. They should not establish their extra accounts and "small exchequers (Xiao Jin Ku)", especially for the fiscal departments. The State Council decided that from 1996, thirteen governmental funds (and fees) which are of relatively large volume should be transferred into the fiscal-budget management. Revenue from the funds (fees) should be delivered to the central treasury or local treasuries in time according to the current system. The line departments are responsible for putting forward the plans and the fiscal department shall make the appropriations according to the relevant regulations and rules. Should it be for capital construction purposes, the fiscal departments will arrange the expenditure according to the project plan approved by the planning departments. It should be managed with two separate streams of revenue and expenditure. The fiscal and audit supervision function should be strengthened. Revenue and expenditure of the funds (fees) should be listed separately in the budget and used as designated. Revenue and expenditure of the funds (fees) should not be used for other purposes, nor should the funds be used for the purpose of balancing the budget. The surcharges and fees collected by the local government in accordance 96 ANNEX 5 with the state regulations will be included in the local fiscal-budget as the fixed revenue of the local fiscal departments and no longer treated as extra-budget funds. Based on the stipulations in the General Financial Rules for Business Enterprises and Accounting Standards for Business Enterprises, the retained funds after tax for the state owned enterprises should no longer be managed as extra-budget funds. Revenues from business and services through market by institutional organizations and social communities, which do not reflect the government functions, are not managed as extra-budget funds. Its revenue may be kept instead of delivered to the special account of fiscal departments, and is subject to tax according to the law, and shall be listed in the financial revenue and expenditure plan of the unit, and be put into the unified accounting for the revenue and expenditure. The State Council asked in the decision to strengthen the management of fees and funds, and strictly control the scale of the extra-budget funds. Collection and withdrawing the extra- budget funds should be done according to the law, regulations and the items, scope, standard, and process that are defined by the legal regulations and rules. Collection of fees for administration and undertaking development should strictly follow the management rules of getting the approval from both central and provincial government. Fee collections should be, according to the subordinating relations, respectively reported to and be approved by the State Council or the fiscal department and the planing (price) department of the peoples' government of provinces, autonomous regions and municipalities. When defining and adjusting the fee collection standards, it should be reported to and approved by the State Council or the planning (prices) department and the fiscal department of the peoples' government of provinces, autonomous region and municipalities. The State Council or people's governments of provincial level should approve import fee collections and the standard formulation as well as the adjustment. Fee collections for administrative undertakings and the collection standards approved by provinces, autonomous regions and municipalities should be reported to the MOF and the SPC for record. All levels of people's governments under provinces, autonomous regions and municipalities (including cities directly under the state budget) and their departments have no right to approve the establishment of the administrative undertaking fees or adjustment of the collection standard. Central government will approve the managerial fees in the administrative fees, fees related to the resources, national license fees and public undertaking fees. Collections of various kinds of administrative undertakings, as well as the local fees related to the central government and other localities that have not been reported and approved, or are not in accordance with the regulations, are categorized arbitrarily as collecting fees and must be terminated. Government fund collections should strictly follow the regulations defined by the State Council and should be reported to and approved by MOF. The important ones should be reported to and approved by the State Council. Application and approval of setting up the funds should be made in accordance with the state laws and regulations and the regulations cited in the relevant documentation of the central government and the State Council. Otherwise it is not allowed. Local governments have no right to approve the setting up of funds projects, nor to approve, in other forms, the setting up of funds in the name of the administrative undertaking collections. The State Council stressed in the decision that the extra-budget funds are from state fiscal funds. They are not the self-possessed funds of the departments and units. They must be put into 97 ANNEX 5 fiscal management. Fiscal departments should open a unified special account in banks for the management of revenue and expenditures of the extra-budget funds. Revenue from extra-budgets of the departments and units must be delivered to the fiscal special accounts of the same level. Expenditures shall be arranged by the same level of fiscal department according to revenue and expenditures of the extra-budget funds plan and the units financial revenue and expenditure plan, and be allocated from the special fiscal accounts, thus to realize the management of the two separate streams of revenue and expenditures. Fiscal departments should establish a management system for the auditing of the extra- budget funds. All departments and all units should compose, according to the regulations, their revenue and expenditure plan of the extra-budget funds and report to the same level of fiscal department, thus to have a unified accounting and unified management of the budgetary appropriation and extra-budgetary revenue. The extra-budget expenditure should be managed carefully. It is forbidden to use it against the regulations. All departments and units should strictly follow the state regulations and revenue and expenditure plans of the extra-budget funds defined by the fiscal departments as well as the revenue and expenditures in the unit financial plans when using the extra-budget funds. Funds and fees targeted for public utilities and public undertakings and other special funds should be earmarked for special purposes according to the plans and regulated usage. They should be checked by the fiscal departments and allocated by installments. Expenditures for paying wages, bonuses, subsidies, allowances and welfare should be made strictly in accordance with the items, scopes and standards defined by the fiscal departments. When making expenditures for the fixed assets investments, the projects should be made in accordance with the state regulations and put into state plans for the fixed assets investments. The expenditures should be made by installments according to the state investment plans and projects' schedules defined by the planning departments. Expenditures for purchasing the specially controlled goods need to be reported to and approved by fiscal departments. Then there is a need of going through formalities for the controlled purchase approval according to the state regulations. It is forbidden to transfer the extra-budget funds to non-financial institutions to manage, set up extra accounts, set up small exchequers or deposit public funds as private ones. It is forbidden to use extra- budget funds for investment in real estate etc., for trading activities in stocks and dealing in futures, as well as various kinds of high level consumption. Fiscal departments should take responsibility to establish and improve various kinds of management systems and actively provide services. They should appropriate extra-budget funds in time and strengthen the management of extra-budget funds. The State Councils, in the decision, also required localities and departments to establish and improve their supervision and penalty system. All levels of people's governments should receive supervision of their same level of people's congresses on the utilization of the extra- budget funds. All levels of fiscal departments should strengthen the management of the revenue and expenditure of the extra-budget funds, establish and improve their inspection system for various kinds of fees and funds, and, together with the PBC, manage well the establishment and management of the accounts of the extra-budget funds. Those who are against the regulation of the extra-budget funds management shall be penalized according to the state laws and regulations. All levels of government should pay attention and strengthen the management of the extra-budget funds. Efforts should be made to clear and rectify the extra-budget funds. If the 98 ANNEX 5 funds are in the category of budget management that are stipulated by the state, they should be included in the budget. If the fees and funds are found to be set up against the state regulations they should be abolished. 99 ANNEX 6 ANNEX 6: LINKING PLANNING, POLICY AND BUDGETING IN A MEDIUM-TERM FRAMEWORK Fiscal crisis, importantly reflecting inadequate links between planning, policy and budgeting, was the genesis of much of the radical restructuring of decision making and budget practice in most OECD countries over the past 15-20 years. It is arguable that the same inadequacies may be the single most important factor contributing to poor budgeting outcomes at the macro, strategic and operational levels in developing countries and economies in transition. The response in OECD countries and in an increasing number of developing countries has been to focus more on a top down approach to budgeting, which acknowledges budget realities at the outset, and to set decision making in a more medium-term framework. Recognizing the tremendous pressures on its budget, China has taken the first tentative steps in developing a more medium-term perspective to its budgeting. This note provides guidance on approaches to improving these linkages throughout the full cycle of planning, policy and budgeting (Figure A6.1). It concludes with an outline of the Australian medium-term expenditure framework (MTEF). Figure A6.1: Linking Planning Policy and Budgeting (1) REVIEW the previous planning and implementation period (6) (2) EVALUATE and AUDIT UNDERTAKE PLANNING activities' effectiveness and feed ACTIVITY the results into future plans Establish resource framework, set out objectives, policies, strategies and expenditure priorities (5) (3) MONITOR activities and MOBILIZE AND ALLOCATE ACCOUNT for expenditures RESOURCES Prepare budget (4) IMPLEMENT PLANNED ACTIVITIES Collect revenues, release funds, deploy personnel, undertake activities 100 ANNEX 6 Integrated planning, policy and budgeting is fundamentally about having expenditure programs that are driven by policy priorities and disciplined by budget realities. The challenge is to manage the tension between "needs" and "availabilities" more effectively (Box A6.1). A medium- term approach provides a linking framework and facilitates the management of the tension between policy and budget realities to reduce pressure throughout the whole budget cycle. The result is better control of public expenditures and better value for money within a hard but predictable budget constraint. Linking planning, policy and budgeting can be accomplished progressively at a pace that fits a country's capacities. The sector is one level at which the process can begin. Many of the potential gains at this level, however, will not be realized if the central planning, resource allocation and budgeting system does not change to support a better balance between policies and resources at the sector level. It will be important to begin to move quickly to implementing a medium-term perspective to central resource allocation and budgeting in order to provide a context for the sector and a government-wide mechanism that forces politicians to confront trade-offs between sectors and policies at a strategic level. Whatever the form of MTEF, it will take a number of years to develop, as it will need to encompass all expenditures and have the following elements: Box A6.1: Needs Versus Availability In many countries, budgeting has been undermined by the "needs" rather than "availability" psychology of the budget actors. While finance ministries stress "availability" (the revenues expected to be forthcoming from domestic and external sources), line ministries persist in basing budget proposals on "needs." The result is a negative-sum budget process that undermines macro-economic stability and program and project effectiveness. These conflicting perspectives are understandable. The finance ministry is constantly battling weak domestic revenues, the debt servicing burden and mounting claims on-budget resources. Since it has responsibility for the macro- economic consequences of fiscal deficits, the MOF has no alternative but to preach the "availability" message through tight budget ceilings. If a satisfactory budget cannot be prepared, it remakes the budget during implementation through the release- of-funds mechanism, at a high cost to project and program efficiency. Line ministries inherit commitments they lack the resources to meet, yet must daily confront the consequences of inadequate allocations. Population growth, the running costs of newly-completed projects, and unforeseen crises further pressure slender budgets. Budget bids are submitted two to three times over finance ministry ceilings. When requested to concentrate resources on priority programs, departments answer they have no mandate to cut service. Privately, they fear that to submit a lower bid within the ceiling would make them vulnerable to further cuts. The debate about "needs" makes no connection with using existing program allocations more efficiently. In countries where the gap between commitments and resources is so large, a satisfactory reduction is unlikely to come about within the tight perspective of the annual budget. A more fundamental public expenditure planning process has to take place engaging finance, planning and line ministries and Cabinet (because inter-sectoral trade-offs have to be faced) in a process that has a medium-term perspective. Donors, too, need to be involved because of their heavy contribution to budget financing. "Availability", both domestic and aid, must be projected beyond a single year and indicative shares communicated to departments. "Needs" must be prioritized and made consistent with the resource framework. All this must happen in a manner that encourages departments to re-examine commitments and their resourcing with central ministries in a more cooperative framework. 101 ANNEX 6 * aggregate fiscal targets * forward estimates of the cost of existing policy * institutional mechanisms for making the trade-offs between "availabilities" and "needs" in such a way that performance is improved at the macro, strategic and operational level * enhanced predictability. Sector Level Planning, Policy and Budgeting Years of short-term planning for annual budgets and hand-to-mouth adjustments during the budget year have led to accumulated overcommitments and inefficiencies at the operational level. The requirement now is to create enough certainty so that line ministries and agencies can plan ahead, have the incentives to do so and have better information on which to base strategic and operational decisions. This requires the development of a comprehensive MTEF. However, integrating planning, policy and budgeting at the sectoral level will provide the building blocks for such a system. The focus should be on the whole sector-a coherent set of policies, programs and activities that need to be looked at together. The objective must be to have a balance between policy and resources, both in the short run and over the medium-term. An early question is whether government has a policy responsibility at all or whether policy should be left to the private sector and/or civil society. Having decided that there is a policy issue for government and that the appropriate instrument is the budget, the possible expenditure implications of the policy issue need to be assessed. A strategic issue that will then need to be confronted is the role of cost recovery in pursuing the policy objective within a constrained budget. The assessment might reveal inconsistencies between declared priorities and actual expenditures. Ideally, decisions should be based on the relationships between input costs, outputs and outcomes. In practice, even basic information on the costs of inputs is often lacking. In most countries, little attention has been given to measuring outputs and outcomes. Poor information makes it impossible to evaluate cost-effectiveness. A good starting point would be to work out what existing activities would cost if fully funded, i.e., if facilities were adequately maintained, if staff were paid a reasonable minimum salary and if essential complementary inputs were provided to help determine whether government is attempting more than it can afford. Mechanisms must be in place to facilitate a shift in resources fiom lower to higher priorities. This means considering elimination of whole programs and activities as well as making "policy" adjustments, e.g. higher pupil teacher ratios. 102 ANNEX 6 Medium-Term Expenditure Framework An MTEF is a centrally-driven strategic policy and expenditure framework within which ministers and line ministries are provided with greater responsibility for resource allocation decisions and results. The key to a successful MTEF is that institutional mechanisms exist which assist and require relevant decision makers to balance what is affordable in aggregate and the policy priorities of the country. It consists of a top-down resource envelope, a bottom-up estimation of the current- and medium-term costs of existing policy and, ultimately, the matching of these costs with available resources. The latter should normally occur in the context of the annual budget process, which should focus on the need for policy change to reflect changing macro-economic conditions as well as changes in strategic priorities of the government. Conservatively, defining medium-term resource envelopes for departments should help change the psychology of budgeting from a "needs" to an "availability" mentality. An MTEF is particularly relevant to countries that need to restructure public expenditures and aims to: * improve macro-economic balance by developing a consistent and realistic resource framework; * improve the allocation of resources to strategic priorities between and within sectors; * increase commitment to predictability of both policy and funding so that ministries can plan ahead and programs can be sustained; * provide line agencies with a hard budget constraint and increased autonomy, increasing incentives for efficient and effective use of funds. Macro-Analysis and Modeling Linking economic projections to fiscal targets. The transition from planning to budgeting often suffers from inconsistencies such as overcommitment. This occurs when decisions do not take into consideration opportunity costs and knock-on effects. Models can assist in identifying problems by checking the internal consistency of proposals and by generating accurate forecasts. Models can also illustrate the trade-off between alternative uses of resources and can make explicit the underlying assumptions about relationships and priorities. Constructing a model can expose differences in assumptions about what drives decisions or relationships and reveals deficiencies in data. Personal computers and software have increased the scope for using models for analysis and explanation. Constructing and using models. The value of model building stems from involving the interested parties in reviewing data, discussing different perceptions about the relevant relationships and identifying data requirements. Strategic Framework A high-level consensus is essential to ensure discipline in adhering to expenditure targets and to the agreed procedures for adjusting them. The consensus needs to include political and technical levels and to cover: * broad objectives of policy and the role of government in the economy, * the need for discipline in macroeconomic management, 103 ANNEX 6 * targets for broad aggregates of public revenue and expenditure, and * procedures for setting and revising the expenditure framework and the responsibilities of key agencies. Procedures. The sequence in which the cabinet considers expenditures is important. Decisions on aggregate expenditures and revenue targets and on broad principles of government policy should precede the consideration of detailed expenditure proposals. Forward projections should be prepared at the policy and program level and should be reviewed before the next annual budget is considered in detail. Once high levels agree upon any changes in key policies and broad allocations, lower levels can be entrusted with the details. Appropriate intervals must separate the stages. Line agencies should receive broad guidelines in time to reflect them in their proposals, otherwise strategic issues are set aside in the rush to complete the detailed annual budget. The central agencies need to prepare the initial framework paper, which would review the economic and fiscal situation and prospects, propose aggregate fiscal targets for the planning/budget period, identify key strategic and policy issues and propose sector allocations for line agencies to use in framing their own proposals. Coordination is needed where responsibilities are divided between finance and planning ministries to generate joint submission to the Cabinet at the beginning of the cycle and to collaborate in the review of line agency proposals. All parties need to understand the need for: (a) persistence through successive annual cycles, and (b) a strengthening of the financial administration and information systems and (c) parallel efforts to strengthen macro and sector analysis. Strategic Allocation Sector resource envelopes. Top-down sector resource envelopes with a medium-term horizon are a basis for predictability so that appropriate strategic and efficient operational decisions can be made. Establishing a sustainable macro ceiling for government expenditures over the medium-term, then breaking it down can derive sector resource envelopes. A division between discretionary and non-discretionary expenditures should be made. What is non-discretionary is often exaggerated. A medium-term perspective increases the scope of effective discretion, e.g. over staffing levels and salary obligations. An unallocated contingency can be withheld to cope with uncertainties and to allow for adjustments for unanticipated expenditures, but this should be kept to a minimum, as it can easily become a "slush" fund. The political nature of resource allocations makes it wise to reach agreement on the criteria to be applied to allocations. Agreement on criteria provides guidance on how to adjust to new or altered circumstances and can increase discipline and predictability. It is also important to have reliable and timely estimates of the costs of current government policies because, in aggregate, they indicate the nature and extent of the pressure on the budget and signal the costs of all the previous decisions made by government. Their value to government decision making will depend on their reliability in each of the forward years. 104 ANNEX 6 The following can contribute to the setting of broad allocations: * identifying whole categories of public expenditures that should be phased out; * analyzing the functional classification of expenditures to identify inconsistencies between actual allocations and roles for the public sector; * checking international comparisons for expenditure ratios for each sector; * analyzing underfunding/overcommitment, including composition of expenditures, e.g., the balance between personnel and operating expenditures, trends in real levels of salaries and of aggregate funding, and the extent to which the investment budget has become a disguised vehicle for recurrent expenditures; * considering cost recovery; * reviewing recurrent cost implications of capital expenditures; and * looking for explicit or implied expenditure commitments not already factored into projections. Sector envelopes reflect both the inertia associated with existing policy and the dynamic of the changing priorities of government-political, economic and social. The envelopes should be defined in a way that provides incentives for trade-offs between policies and programs at the sector level. Aid and local funding, and capital and recurrent expenditures should be incorporated within a single guideline. This comprehensive approach should result in pressure to unify the budget and to reduce bias in the choice between investment and recurrent expenditures. This comprehensive approach will take time to implement. The Stages of the MTEF Thus, the stages in preparing an MTEF include: a. developing a consistent macro-economic framework and, from this, realistic resource projections for three years; b. deducting statutory payments such as pensions and debt payments; c. breaking down the remaining aggregate expenditure figures into sectoral ceilings, based on the relative importance of each sector; d. providing line agencies with ceilings for the budget year, together with indicative ceilings for the two outer years; e. determining for each sector which progmms should receive a greater or lesser share of the sector resource envelope (ministries need to determine the actual costs of these activities before they can make realistic and rational decisions about the allocation of resources); f. approval by Cabinet of the annual budget; g. preparation of annual budget estimates (year one of the MTEF); and h. approval of the budget by Parliament. 105 ANNEX 6 Factors that determine the success or failure of an MTEF include: * political commitment and endorsement at the highest level to make and abide by the difficult decisions involved in the restructuring of expenditures (some ministries may need to scale back their activities so that more resources can be directed to higher priority sectors); * understanding of, and commitment to, the difficult decisions at the sector ministry level; * commitment at all levels to abide by the budget decision so that new expenditure decisions are not introduced during budget implementation that would require reallocation of resources (these new decisions mean that the priorities set when the budget is approved by Parliament are often overturned); * improvements in expenditure control so that the decisions are not undermined by overexpenditures and reallocation of funds during budget implementation; * improved macroeconomic management and revenue collection so that revenue shortfalls do not necessitate adjustments to the budget estimates; * briefings of politicians and senior management during implementation; * improvements to expenditure reporting on results; and * development of a computerized accounting system. The MTEF provides a framework within which strategic sector reviews should be undertaken (Box A6.2). This takes us a full circle back to the comment at the beginning of this note about the importance of setting the linkage of planning, policy and budgeting within the context of an MTEF. The development of an MTEF is very much an iterative process, as is the annual budget process once an MTEF is in place. Stage 1 involves developing the expenditure framework, taking account of resource availability, the role of government and priorities between sectors (the iterative nature of the process is highlighted here, as establishing priorities between sectors requires the information generated by stage 2). Stage 2 involves the sector review process, through which each sector (including all ministries and institutions) would define objectives, policies and strategies, together with the costs of current and proposed sector policies over the medium-term. These would need to be sufficiently clear so as to enable the ministries and institutions to define the functions and activities involved in implementing the agreed strategies. Stage 3 involves identifying structures and personnel required, the recurrent and capital costs of implementing the activities, and the projects associated with these costs. Stage 4 involves monitoring staff performance, costs and program performance. 106 ANNEX 6 Box A6.2: An Integrated Strategic Planning Process STAGE 1: Medium-Term Expenditure Framework Macroeconomic Framework Resource Framework Role of Government Sectoral Ceilings STAGE 2: Strategic Sector Objectives, Policies, and Strategies Functions and Activities STAGE 3: Ministry Structures Expenditure Requirements Personnel Recurrent Costs Capital Costs Requirements Project Costs STAGE 4: Monitoring Information Personnel Accounting Information Performance Indicators Performance Monitoring 107 ANNEX 7 ANNEx 7: THE AUSTRALIAN EXPERIENCE WITH A MEDIUM-TERM EXPENDITURE FRAMEWORK When Australia embarked on its comprehensive reform program in the early 1980s, a key consideration was the perceived inadequacies in the links between policies and programs and the resources allocated to their implementation. Fiscal crisis subsequently raised fundamental concerns about the affordability of current government policies. The response to this was to take the system of forward estimates that had played a peripheral role in decision making and place it at the center of both resource allocation decision making and resource use. Aggregate Fiscal Targets Beginning in 1985, the central government adopted a medium-term "trilogy" strategy of not increasing outlays or revenue as a proportion of GDP and of reducing the deficit/GDP ratio. Subsequent economic crisis saw this commitment tightened to no real increase in expenditures. The credibility of the forward estimates discussed in the next section was central to the success of this strategy. The end of the 1980s had been converted to the deficit of 4 percent a surplus of 2percent; government had significantly reoriented expenditures to reflect its core strategic priorities and the incentives for efficient and effective use of resources had been considerably strengthened. Recession in the early 1990s saw a return to deficits and fiscal targets have been focused on a realistic time path for returning to balance (the 1997-98 budget deficit is forecast to be less than one percent). More significantly has been the "Budget Honesty" commitment of the government that requires the government to regularly publish projections of expenditure and revenue, notably in the three months prior to an election. (The New Zealand Fiscal Responsibility Act goes even further by committing government to make public its long-term fiscal objectives and to pursue policies which are consistent with maintaining crown debt at a prudent level and with a reasonable degree of predictability about the level and stability of tax rates in future years.) Forward Estimates of the Cost of Existing Policy The "forward estimates" system evolved in Australia in the late 1970s through the 1980s. The forward estimate process develops estimates that, on a rolling basis, project the level and composition of expenditures for three years beyond the current fiscal year, assuming no policy changes. These are adjusted requlary to take account of factors such as inflation, where program expenditures are indexed, and government policy decisions that may increase or decrease estimated costs. The practice prior to 1983-84 involved the Department of Finance collecting bids for program spending from sponsor departments without rigorously examining the basis for them, except with respect to the first year. Accordingly, these bids reflected departments' own assessments of their future needs, a practice described by M. Keating and M. Holmes (1990) as "a major cause of creeping incrementalism of government [expenditures]." 108 ANNEX 7 Under the new approach, the Department of Finance negotiated with departments the estimates for existing programs and then assumed responsibility for updating the forward estimates at regular intervals to reflect, as indicated above, changes in economic parameters, other technical variations and, most importantly, the effects of government policy decisions. The same process is followed with new policy and program proposals, for which projected costs for the full forward estimates period are required as part of the policy proposal considered by Cabinet. Thus, the Department of Finance is seen as "owning" the forward estimates. Furthermore, whereas previously there tended to be widespread annual renegotiation of estimated expenditures, the new system is much more policy focussed, involving ministers primarily in the relatively small percentage of budgetary matters that require policy or strategic decisions (although the funding implications may involve a high proportion of budget funding). Thus, the forward estimates are a disciplining mechanism in the budgeting process that enables a greater focus on strategic policy issues. At the same time, they provide much greater predictability as to resource levels for departments and agencies. In essence, the system envisages that if government policy does not change then funding will be provided in accordance with the forward estimates. One senior offical has attested to the significant impact of forward estimates as follows: The fact that we now have a budget system in place with forward estimates and that the haggle over the base for each new budget year does not take place any more is a huge advance. If you had to pick out the one thing that we have done above all others, this reform would be the most dramatic change. In 1983, a significant decision in the evolution of the forward estimates system was made when the government decided to publish them. The requirement to disclose costs for the three-year forward period was intended to ensure that decisions were made with greater awareness of future commitments and to provide Parliament and the public with better information about budgetary realities and public expenditure patterns and priorities. The decision to publish also meant that forward estimates had to be taken more seriously, thus leading to their progressive upgrading (M. Keating, M. Holmes 1990). As the system has evolved, the government is required to disclose and justify the costs of policy decisions leading to discretionary changes in expenditures over the three-year forward estimate period. The estimates are published in the budget alongside the budget year figures and changes between the forward years and the budget are reconciled in budget documents-that is, the budget estimates are reconciled with the forward estimates compiled the previous year. These reforms have tended to shift the focus for ministers and senior officials to a medium-term period (of four years), rather than the current budget year. The impact of the forward estimates has been such that an evaluation of government reforms in 1993, in linking the forward estimates system to the record of overall government spending restraint, characterized them as central to the expenditure control process." More recently, a former official who had been instrumental in the development of the government's budgetary and financial management reforms of 1980s stated: 109 ANNEX 7 The forward estimate process and system was so central because it provided the backbone which linked the Expenditure Review Committee's (see below) macro- economic and strategic policy-making, portfolio budgeting and the running costs system [the latter is discussed in a separate note]. It has provided a framework for a more strategic approach to decision making, much greater predictability in funding for current policies and for removing from the budgetary arena those decisions best made elsewhere (most notably management decisions). The system has built on trust and has changed behavior fundamentally. Perhaps the most important factor here has been the fact that, having changed the formal rules, all the players have played by the new rules (1996 interview). Institutional Mechanisms for Making the Trade-off There can be little doubt that the Expenditure Review Committee (ERC) established by the Labor Government in the mid-1980s was central to the subsequent improvements in all three levels of budgetary outcomes. This committee was a sub-committee of the Cabinet, consisting not only of the Prime Minister, the Treasurer and Minister of Finance but also of a number of other senior "spending" ministers. This committee was responsible for determining the overall fiscal framework and for managing strategic policy making, including policy changes necessary to reflect fiscal realities as well as the shifting priorities of the government. One of the key strategic decisions made by the ERC was the resource envelope for each sectoral minister for finalization of the annual budget. Depending on whether the envelope was higher or lower than the forward estimates of existing policy (adjusted for the individual policy decisions made by ERC - see previous paragraph), individual sector ministers would have to seek programmatic changes that would produce savings or they may be able to introduce new initiatives. The key point here is that it was left to sector ministers to determine the best allocation of resources to policies and programs in their sector consistent with overall government policy and within a hard budget constraint. The third element of the system was the running costs system. This system, which is discussed in some detail in a separate annex, provides line managers with considerable flexibility in managing their personnel and administrative resources within a hard budget constraint but one that is predictable over the medium term. This system has eliminated the annual haggle over funding levels for administration and has meant that ministers have been freed from involvement in decisions at this level. It is the efficiency dividend component of the system that has enabled decisions on running costs to be kept out of the Cabinet arena and has built the trust between line agencies and the MOF. Finally, it is worth noting that the forward-estimates system has Umabled the MOF to assume something of a banker role. This can be illustrated by reference to the example of the major modernization of the Australian tax system. This involved an investment of over A$1billion. Because of the forward estimates, MOF was able to reduce the Tax Office's running costs in the outyears for the savings generated by the investment. The benefits from this are obvious, not least in the changed incentives on line agencies to have sound analysis of expenditure proposals. 110 ANNEX 8 ANNEX 8: THE RUNNING COST SYSTEM IN AUSTRALIA The Running Cost System is an integral part of the broader Public Expenditure Management System, which is designed to provide managers with: a) Substantial flexibility in the use of budgeted running costs and greater predictability over resource levels beyond the budget year, and b) A strategic framework, with associated evaluation and reporting requirements, which encourages and requires managers to focus on results and the resources required to achieve them. The objective of the running cost arrangements is to improve the efficiency and effectiveness of resource use in the public sector. The arrangements were introduced in 1987-88 as part of the Central government's Financial Management Improvement Program. The arrangements: * Provide managers with increased authority and responsibility for the use of resources; * Provide incentives to agencies to improve public sector efficiency and effectiveness by allowing them to retain the majority of savings arising from increased efficiency; * Restrain costs growth through cash limiting and restricting access to supplementation; and * Make efficiency gains visible by applying an annual efficiency dividend. Definition of Arrangements The running cost of an agency represent the full recurrent costs, and minor capital costs, consumed by an agency in providing the Government services for which it is responsible. * These include salaries and related employment costs, superannuation, administrative items, minor capital items, property operating expenses and the purchase of goods and services from the private sector including consultancy services, research, advertising and public relations services. * These exclude all transfer payments (pensions, grants, subsidies), major capital payments and loans. The running cost arrangements are a set of rules and procedures, which enable agencies to make most efficient use of the resources available for running cost. The running cost arrangements apply to budget funded agencies operating on the Commonwealth Public Account. Legislative framework The Constitution provides that funds may not be spent unless appropriated by Parliament. The running cost arrangements are therefore implemented within the legislative framework of the Commonwealth Budget. 111 ANNEX 8 The running cost budgets of agencies subject to the running cost arrangements include funding from: * Parliamentary appropriations: - included in the Appropriation Act (No. 1) and Supplementary Appropriations (Additional Estimates) or equivalent appropriation Acts for the parliamentary Departments; and * Retention of receipts through section 35(3) of the Audit Act 1901 (see below). Notional Items A notional item sets a maximum limit on the amount that can be spent on a component within the appropriation. From 1995-96 there are two notional items for each functional classification within the running cost appropriations for most agencies: * Senior Executive Service (SES) Salaries; and * Other Running Costs (i.e. including all non-SES salaries). Running Costs Flexibility Under this structure, agencies have full flexibility to shift resources within other running cost to meet competing priorities without needing to seek the agreement of the Department of Finance. The notional items are the only points of control on expenditures within the total cash limited running cost budgets. Transfers between running cost and non-running cost appropriations require the approval of the Secretary of the Department of Finance. Cash Limiting Annual running cost budgets are cash limited. * The onus is placed on agencies to manage within the confines of a set running cost budget. * Agencies must absorb the entire running cost associated with new policy decisions outside the budget context. Costing of variations to running cost estimates will generally not be reopened for further consideration except by reference to the Cabinet in the new policy context of the annual budget. Agencies must absorb all or part of any increase in running cost sought for new policy except when the total portfolio new policy is fully offset by agreed savings in program or running cost or by agreed revenue increases. The amount to be absorbed is equivalent to one percent of the existing running cost base of the agency before the addition of new policy, savings and one-off adjustments. 112 ANNEX8 Price Adjustment Administrative Expenses and Property Operating Expenses These items are indexed annually by different deflators-administrative expenses (including legal services) are indexed by the Commonwealth Final Government Consumption Expenditure deflator; and property operating expenses are indexed by the Running Costs Property Index. Indexation is finalized prior to each budget. Each item is indexed by the latest available deflator outcome, that is the actual price movements for the previous calendar year. This replaces the previous arrangements where items were indexed to forecast changes and subject to ex-post price adjustment. Agencies now have funding certainty from the commencement of the financial year and are able to plan accordingly. Salary Supplementation If Cabinet approves Supplementation, SES and non-SES salary estimates are supplemented for award increases and other approved wage and salary related decisions. Efficiency Dividend The efficiency dividend was introduced in 1987-88 as an integral component of the running cost arrangements. All agencies must pay the efficiency dividend unless specifically exempted by Cabinet. The efficiency dividend aims to: * Provide managers with the incentive to continually seek new or more efficient means of undertaking ongoing government business; * Allow government to redirect a portion of efficiency gains to higher priority activities; and * Clearly demonstrate Public Service efficiencies resulting from improvements in management and administrative practices. Statutory bodies not operating on the public account are also required to pay an efficiency dividend, unless specifically exempted. The current efficiency dividend rate applying to running cost agencies is one percent per annum (previously 1.25 percent). This is applied to the total net running cost (i.e. excludes receipts through section 35 net annotated appropriations). The efficiency dividend is applied before the running cost base is indexed to budget year prices and before any other one-off ongoing adjustments are made. 113 ANNEX 8 Resource Agreements A resource agreement between an agency and the Department of Finance is an agreement for the provision of resources in return for some action, undertaking to act, or some other consideration. It provides the agency with additional resourcing flexibility outside the confines of the regular budget process in order to achieve some stated objective. Common types of agreements include: * Receipts retention and sharing (including section 35 agreements); * Multiple year carryovers and borrowings; * Carryovers of greater than 10 percent and borrowings of greater than 6 percent; * Workload adjustment formulae; * Property resource agreements; and * Workplace bargaining agreements. Carryovers Carryovers are the mechanism by which resources appropriated in one year can be spent in some future years. Carryovers: * Provide agencies with the flexibility to respond to changing spending priorities; * Reduce the incentive for unnecessary end-of-year spending by agencies; and * Allow managers to plan spending without being restricted by cut-off dates. Up to 10 percent of the total running cost budget (including section 35 annotated appropriations) can be carried forward from any one year. Multiple Year Carryovers A multiple-year carryover is a carry forward of funds over at least two years where the funds carried over are not used in the intervening year or years. Multiple-year carryovers are subject to a resource agreement. The resource agreement should state: * The specific purpose for accumulating funds; * The interest rate to be applied (currently 8 percent per annum); and * When the funds are expected to be required. Interest is payable on multiple-year carryovers which are negotiated in advance with the Department of Finance. Interest is paid initially at the budget following the end of the financial year from which the carryover was made. Interest is then paid at every following budget (i.e. in May after previous June end of financial year) on the cumulative carryover amount for each year the carryover is maintained. 114 ANNEX 8 Borrowings Borrowings allow agencies to bring forward some future years' appropriation to be spent in an earlier year. This provides agencies with: * The flexibility to respond to changing spending priorities; and * A mechanism to meet unforeseen costs. Borrowings can be arranged either: * Through annual appropriation or additional estimates bills, with a consequent reduction in a future appropriation; or * Through the Provision for Running Cost Borrowings where the borrowing does not coincide with an appropriation bill. Agencies are allowed to borrow up to 10 percent of the total running cost budgets (including section 35 receipts) from any one year. * Borrowings of greater than 6 percent are allowed provided a resource agreement is signed. * Borrowings of over 10 percent are allowed only in exceptional circumstances. Multiple- Year Borrowings Borrowing running cost over multiple years is allowed provided a resource agreement is signed. The resource agreement should state: * The specific purpose for borrowing funds; * The interest rate to be applied (currently 8 percent per annum); and * When the funds will be repaid. Interest is payable on multiple-year borrowings that are negotiated in advance with the Department of Finance: Interest is paid initially at the budget following the end of the financial year into which money was borrowed. Interest is then paid at every following budget (i.e. in May after previous June end of financial year) on the amount still owing for every year money is owed. Retention of Receipts Section 35 of the Audit Act 1901 has the effect of annotating agency appropriations by the amount of the receipts raised. These receipts form part of the running cost of agencies. Agencies may enter into an agreement with the Minister for Finance or his/her delegate and retain proceeds from: * The sale of under-performing non-property assets; * The sale and rental of staff housing; * Proceeds from user charging for the provision of services (refer to Section below on User Charging); 115 ANNEX8 * Sub-leasing of property or resale of goods used in fitting out premises, and other minor categories of receipts. Section 35 Agreements should: * Identify sources of funds; * Specify estimates of receipts; * Specify sharing arrangements (if applicable); * Specify the financial period for the agreement; * The accounting bases (cash or accrual); and * Any specific reporting requirement. Section 35 agreement must be renewed (or extended) every time an appropriation bill before Parliament references the arrangement (i.e. at budget and additional estimates). Renewals and extensions must: * Make reference to the original agreement; * State the intention of the parties to extend the time period or amounts detailed in the original agreement; * State whether the conditions contained in the original agreement apply or detail any variations(s) to the original agreement; * State the period for which the renewal is effective; * State the estimated receipts for the period; and * Be signed by appropriate department or agency delegates. Agreements must be revised if the categories of receipts to be covered or the estimated amount of receipts changes. User-Charging Receipts User-charging activities are those where all or part of the cost of providing goods or services is met through charges paid by users rather than through direct budget funding. For running cost agencies, user charging is funded through: * net appropriation arrangements under section 35 of the Audit Act 1901, or * trust account operations where the charter does not provide for fully commercialized activities. Agencies can increase user-charging receipts for any activity by up to $1m per annum without an adjustment to the level of funding provided by the budget. An increase of over $lm will result in any budget funding for the cost of that activity being withdrawn unless otherwise agreed by the Minister for Finance. 116 ANNEX8 Agencies may retain all receipts associated with user charging where annual turnover (measured on a predetermined cash or accrual basis) are less than $1Om. If turnover is more than $1Om, receipts are to be shared if there is a surplus. * A surplus arises where receipts exceed the cost incurred by an agency in producing the goods or services that are charged for. * If receipts do not exceed the cost of provision, all receipts are retained by the agency. * If there is a surplus of receipts over costs, it will be shared between the agency and the budget. Sharing of surplus will usually be on a 50:50 basis. The sharing arrangements are to be applied on an agency basis against the $10m threshold unless the Minister for Finance or an authorized officer agrees to alternative arrangements. Funds collected through net annotated appropriations that do not arise from user charges (for example, staff contributions to benefits, sales of surplus assets) will not be included in the assessment of receipts for revenue sharing purposes. An agency is required to provide sufficient information to the Department of Finance to enable the extent of surplus to be determined. Revenue relating to user charging activity should be identified in the annual financial statements of agencies. In addition, resource agreements may require agencies to provide special purpose reports in some cases. As part of their corporate plan agencies are to develop a business-planning segment relating to significant user charging activities. Significant user charging activities are those which exceed the $1 Om threshold or involve high risk. High Risk Activities The Chief Executive is responsible for undertaking a risk assessment of the activity in association with annual planning arrangements. High risk can include both financial and general dimensions, where a user charging activity may not achieve results consistent with planned objectives or costs, involve potential sensitivity for the government or has potentially significant budgetary or contingent financial liability. The Chief Executive is responsible for advising the portfolio Minister. If the portfolio Minister agrees with the Chief Executive's assessment, enhanced monitoring and accountability arrangements should be established in consultation with the Department of Finance. 117 ANNEX 9 ANNEX 9: CHINA'S INTERGOVERNMENTAL FISCAL RELATIONS As local governments and their NPOs control the bulk of EBFs, implementation of SC 29 will have a direct and important effect on local budgets. By sharply restricting the authority to impose fees and levies, the proposed changes to EBF under SC29 have the potential effect of significantly reducing resources accruing to local government. The problem is especially serious for rural governments (affecting directly township governments, but by extension also county governments). Lacking the authority to impose taxes under the fiscal system, EBF (and self-raised funds) have been used by local governments as de jure taxing mechanisms. This has been tolerated and even condoned by the authorities. For example, while all levels of government depend to varying extents on users' fees, asset sales and other nonbudgetary resources, at the township level there is explicit official recognition of nonbudgetary sources for financing government-since their founding in 1986, township finance bureaus have been required to report receipts and expenditures of "three types of funds": budgetary, extra-budgetary, and self raised funds. A State Council decree in 1991 explicitly permitted township governments to impose a fee on TVEs to finance agricultural subsidies, called the "using industry to aid agriculture" fee. The blurring of budgetary and extra-budgetary concepts at the level of local governments is exemplified by the reference to off-budget funds as "the second budget", whose uses are virtually identical to those of budgetary revenues. They are spent on education, health, farmland investment, road maintenance and construction, family planning services, administration (paying salaries and subsidies for cadres), etc., in ways that clearly supplement the budget. At the subnational level, governments often make explicit transfers from EBF to augment budgetary revenues and to cover budgetary expenditures. Virtually everywhere off-budget resources are the main source of finance for infrastructural investments. In implementing SC29, the government must therefore pay particular attention to revising intergovernmental expenditure or revenue assignments to provide incentives for local compliance and avoid creating new forms of off-budget finance. While it is necessary to curb the proliferation of funds and levies, the MOF must take into consideration that EBF have been an important source of local finance. These funds cannot be simply eliminated or recentralized without imposing increased pressure on local finance. Reform in the management of EBF thus creates an urgent need for accelerating the pace of reform in intergovernmental fiscal relations. An appropriate system of intergovernment fiscal relations (IFR) is critical to achieving efficient public expenditure management in China where the fiscal system is so decentralized. This is because a larger share of expenditures in the hands of local governments can affect aggregate fiscal discipline, as local governments are not concerned with maintaining macro stability; despite legal barriers, local governments sometimes borrow to finance their deficits. The sharing of spending responsibilities sometimes leaves no one in charge of aggregate budget balancing. Moreover, with the central budget accounting for less than 30 percent of expenditures, ensuring that the composition of budget expenditures reflects national or social priorities requires 118 ANNEX 9 that local spending be aligned with these priorities. The present system of IFR is poorly designed to achieve such harmonization, as it provides neither the right incentives nor a good match between expenditure responsibilities and revenue capacity. Through the transition period, IFR in China has been plagued by problems of inadequate revenues as the budget declined as a share of GDP, overlapping and murky divisions of labor whereby several levels of government share responsibilities for investment in economic development, education, and health care; and inadequate equalization transfers that left large disparities in the provision of basic services across regions. At the lower end, the levels of provision for basic education and health fall far below standards set by national policyand their inadequate provision may be preventing the spread of growth and its benefits from reaching the poorest regions. IFR deteriorated as governments at all levels tended to respond to fiscal pressure by attempting to devolve expenditure responsibilities while retaining control over revenues assigned to them. Thus, revenue assignments became increasingly mismatched with expenditure assignments over time. Budgets at the subnational levels were additionally squeezed as rapid wage increases drove up the costs of providing social services and day-to-day administration, which are local responsibilities (Wong 1991). At the county level, wages and running costs of government reportedly absorb up to 80 percent of budgetary expenditures. By the early 1990s, a climate of distrust surrounded intergovernmental fiscal relations. The central government attributed the continuing fiscal decline in part to local government unwillingness to collect taxes while diverting funds from budgetary to extra-budgetary channels. Local governments saw the repeated changes in revenue sharing rules through the 1980s-when four different systems were successively introduced-as a lack of firm commitment by the central government to revenue assignments. The introduction of new taxes (e.g. the Energy and Transport Key Construction Fund and the Budget Adjustment Fund) and several rounds of "borrowings" all helped to convince local governments that surplus revenues were not safe from the central government (Wong, Heady and Woo 1995). The reform package of 1994 was centrally concerned with improving intergovernmental fiscal relations, whose centerpiece was the "fenshuizhi" (FSZ or tax sharing system) reform. The intent of FSZ was: 1) to simplify and rationalize the tax structure by reducing tax types, reducing tax rates, unifying the burden on taxpayers and reducing exemptions; 2) to raise the revenue-GDP ratio; 3) to raise the central share of total revenues; and 4) to put central-local revenue sharing on a more transparent, objective basis by shifting from the negotiated sharing of general revenues to a tax assignment system. Under the FSZ, taxes were reassigned between the central and local governments. Central taxes (or "central fixed incomes") include customs duties, the consumption tax (CT), VAT revenues collected by customs, income taxes from central enterprises, banks and nonbank financial intermediaries; the remitted profits, income taxes, business taxes, and urban construction and maintenance taxes of the railroad, bank headquarters and insurance companies; and resource taxes on offshore oil extraction. Local taxes (or "local fixed incomes") consist of business taxes (excluding those named above as central fixed incomes), income taxes and profit remittances of local enterprises, urban land use taxes, personal income taxes, the fixed asset investment orientation tax, urban construction and maintenance tax, real estate taxes, vehicle utilization tax, the stamp.tax, animal slaughter tax, agricultural taxes, title tax, inheritance and gift 119 ANNEX 9 taxes, capital gains tax on land, state land sales revenues, and resource taxes derived from land- based resources. Shared taxes consist of the VAT, which is shared at the fixed rate of 75 percent for the central government, and 25 percent for local governments; and the new securities trading tax, which is shared equally between the central and the local governments. Since FSZ was implemented, significant improvements have been achieved in the first three objectives. The tax structure is greatly simplified. The VAT has replaced the product tax and has been implemented basically at a uniform rate of 17 percent. The enterprise income tax structure has been merged, so that SOEs and COEs and other enterprises are all on the same tax schedule, and the top rate has been reduced from 55 percent to 33 percent. The enterprise profit and tax contracting has been largely eliminated, hardening enterprise tax obligations. Some success has also been achieved with the second objective of revenue mobilization. In 1991 and 1992, revenue elasticity (percent growth in revenues divided by percent growth in GDP) had fallen to 0.44 and 0.46, respectively. In 1993 it rebounded to 0.83, partly as a result of local governments' efforts to "ratchet up" their bases as the central government promised to rebate sufficient revenues from the VAT and consumption tax to maintain local revenues at the 1993 level. In 1994 and 1995, revenue elasticity was 0.57 and 0.78, respectively (Xu 1997). These were better than the 1991 levels but still below one, so that the revenue-GDP ratio continued to fall. In 1996 this trend was reversed for the first time since the beginning of market reform in 1978, when revenues grew by 18 percent to 736.7 billion Yuan. This trend has continued through 1997 and 1998, with the revenue-GDP ratio rising to about 13 percent, giving rise to some optimism that FSZ is succeeding in reversing the fiscal decline. The third objective of raising the central share of revenues was achieved in 1994, when the one-time transfer of the bulk of VAT and CT revenues to central coffers boosted the central share from 22 percent of total revenues to 55.7 percent. Since then, however, the central share has fallen steadily, to just below 50 percent in 1996, because local taxes have grown faster than VAT and CT revenues. Figure A9.1 shows the changes in the central share of revenues and expenditures since 1978. The mixed effect of FSZ on recent trends is highlighted in Figure A9.2. While FSZ has recentralized revenues, by policy design it has not altered the continuing decentralization of expenditures. Indeed, central own expenditures have fallen from 33 percent of the total in 1990 to 29 percent in 1998. It is in the area of improving intergovernmental fiscal relations that the FSZ reform appears to have produced the most disappointing results. In one sense, this is not surprising: local governments simply do not like giving up "ownership" of revenues and yielding control to the central government. Even after five years, the premise that taxes belong to the central government unless specifically assigned to the localities is widely resisted at the local levels. That central transfers are now accounting for nearly half of total local expenditures means a reduction of local autonomy over allocative decisions. Other factors contribute to the unpopularity of FSZ and the lack of improvement in intergovernmental fiscal relations. One is complexity. By setting uniform rates of sharing across provinces for the VAT, CT and securities trading taxes, the FSZ was intended to make the revenue sharing process more open, transparent and objective. In the process of implementation, several distortions have emerged that have made the system far more complex-the common 120 ANNEX 9 saying is that the system of revenue sharing in China is such that "insiders find it impossible to explain, and outsiders find it impossible to comprehend" (neihang shuobuqing, waihang tingbudong). Figure A9.1: Changing Central Shares Revenue Share (%) aCentral Local 100.0 90.0 80 0 70 0 Local 60.0 50 0 40 0 30 0 20 0 10 0 - ~ Q o c ) 0 Expenditure Share (%) centra 100 0 90 0 80 0 70 0 Local 60 0 50 0 40 0 30 0 20 0 100 0.0 121 ANNEX 9 Figure A9.2: Central Revenue and Expenditure After Local Transfers 1990 Revenue 1996 Revenue Central Own Revenue Local 34% Net Revenue Central NetRevenue Own Revenue Local= 42% 50% Local Local Net Net Remittance Remittance 16% 8% 1995 Revenue 1995 Expenditure Locally Central Financed Own Expenditure Expenditures Local 34% 29% Net Revenue O t 38% Central Own Revenue 52% Local Transfer Net s to Localities Remittance 37% 10% 1996 Revenue 1996 Expenditure Central Own Locally Expenditures Locally27% Net Revue Own Revenue 42% ~ ~ ~~~~50% ~'____________ Local Transfer Net s to Localities Remittance 34% 8% Source: MOF, China Finance Yearbook 1996, p. 535, and Liu Zhongli (1997). The Calculation of Tax Rebates The goals of uniformity and transparency were compromised at the outset when the central government agreed to accept the status quo distribution in 1993. This committed the fiscal system to the "simultaneous use of two systems", with a form of contracting being retained (with quota subsidies to the poor provinces and tax rebates to all provinces to make up the gap between local revenues under the FSZ system and the retained local revenues in 1993), and FSZ being applied only at the margin, to incremental revenues. 122 ANNEX 9 Available local revenue now consists of the following: Local taxes + central transfers = LT + LTincrease + 0.25x (VAT+ VAT increase) + 0.3x [0.75x (VAT increase) + (CT increase)] + contracted tax rebate + other central transfers where: contracted tax rebate = PBR - LT - 0.25 x VAT and PBR is the province's base retained revenue LT is the province's local tax revenue in the base year (1993) VAT is the VAT revenue from that province in the base year CT is the consumption tax revenue from that province in the base year This reduces to: Available local revenue = PBR + LT increase + 0.475 x VAT increase + 0.3 x CT increase + other central transfers Implementation at the subprovincial levels. Even greater complexity is introduced in the application of the FSZ system to subprovincial levels, which differs across provinces. The reform package had specified only the assignment of taxes between the central and provincial governments, and left it to each province to determine how to share the subnational portion among the lower levels. Some provinces have simply passed through the formula for sharing to lower levels, while some provinces and prefectural units have added their own layers to the tax sharing system. For example, in Guizhou, the province takes 10 percent of the locally retained 25 percent of VAT, and 50 percent of the incremental portion of four local taxes-land occupation tax, land appreciation tax, revenues from land leases and the resource tax. In some provinces including Liaoning and Jiangsu, many local taxes are shared between the province and municipalities. The proliferation of shared taxes at the subnational levels not onlymakes revenue assignments to subprovincial units more difficult to calculate, but it also introduces inequalities across provinces that may distort incentives. The Introduction of Revenue Growth Targets The FSZ system was also complicated by the revenue growth targets (kaohe zhibiao, or reference targets) that were introduced by the MOF in 1994, as a retaliatory measure aimed at provinces that were suspected of having artificially inflated revenues in 1993. The announcement in August 1993 that 1993 would be used as the base year for the FSZ reform led to a surge in tax collection during September-December, when turnover tax revenues for the year increased by 49.6 percent over the 1992 figure. It was suspected that many provinces inflated their collections by calling in not only tax arrears but may have persuaded enterprises to pre-pay their 1994 tax obligations. To prevent these provinces from gaining and to ensure revenue growth in 1994 and after, the MOF sets targets for revenue growth for each province, with the threat of reductions in base revenues for those provinces that failed to meet their target. The effect of these reference targets was to increase fiscal pressures on subprovincial governments, and these pressures in turn spawned ever more creative systems of rewards and punishment. Even though MOF has not issued revenue growth targets since 1994, the practice seems to have continued at the subnational levels. For example, Puding County, a nationally 123 ANNEX 9 designated poor county in Guizhou, received a revenue growth target of 20 percent for 1997 from the provincial fiscal department. To meet this target, the county finance bureau has set contracts mandating revenue growth (after tax sharing with the central government) of 15 percent, 20 percent and 25 percent per year with its eleven subordinate townships and towns. Those meeting their targets will be given rewards of Y10,000; 20,000 and 30,000 for the collective, while the township leaders will get Y3-5,000. On the downside, those not meeting their targets will be penalized by amounts equal to the rewards (i.e. Y30-50,000 for the collectives and Y3-5,000 for the leaders), with the fines paid by the collective and the leaders, respectively. Large additional rewards are in store for those meeting their targets three years in a row. This is a structure that tends to exacerbate the good and ill fortunes of the townships and towns. The worry is that the widespread introduction of such incentives may give rise to falsification of fiscal data at the unfortunate localities, where officials will be tempted to report nonexistent revenues on one side of the ledger and inflate expenditure figures on the other side to avoid penalties. The Effect of FSZ on Regional Disparities Also fueling the disaffection for FSZ is the widespread perception that it is exacerbating the unequal distribution of revenues and expenditures. This outcome was anticipated at the outset as FSZ shifted the revenue sharing principle from the previous, weakly redistributive one to the derivation principle-where transfers are pegged to collections derived from the region, mandating that more is returned to the revenue-rich regions (cities) than to poor ones (rural sector). The inability of the old system to redistribute income across regions was well recognized by the early 1990s, when the search began for a new system of intergovernmental transfers that would move decisively away from the historical base numbers to a more objective, formula- driven scheme. The introduction of the FSZ system was accompanied by the strong argument that the central government needs to allocate a larger share of total revenues in order to play a bigger redistributive role. It is in light of that argument that the present intergovernmental transfers are disappointing, as they are dominated by tax rebates, which are strongly dis- equalizing. In the pre-1994 system, central transfers to the provinces consisted of three components: quota subsidies, earmarked subsidies and other subsidies. In line with the "simultaneous implementation of two systems," the FSZ reform has added a fourth component of contracted tax rebates while retaining the other three. Figure A9.3 shows per capita transfers by province in 1996. It shows tax rebates to be highly correlated with per capita GDP while there is no correlation between non-tax transfers and GDP. The bottom panel shows a positive correlation between GDP and total central transfers, attesting to the dominance of tax rebates and the overall disequalizing effect of the present system of intergovernmental transfers. 124 ANNEX 9 Figure A9.3: Per Capita Transfers by Provinces 12 Shanghai * 10 8 6 Beijing x Tianjin * 1-4 Yunnan 2 Guizhou * 4 * **.4 * * 0 30 60 90 120 150 180 210 GDP 18 16 * Tibet 14 12 0 5 10 x 8 -6 Z z 4 Shanghai Tianjin 2 Guizhou Beijing 0 0 30 60 90 120 150 180 210 GDP 18 16 Tibet 14 Shanghai 4 12 10 8 6 Tianjin * Beijing 4, 4Guizhou 2 0 30 60 90 120 150 180 210 GDP 125 ANNEX 9 Illustrative Box on Guizhou Province The effect of FSZ on one of the poorest provinces in China, Guizhou Province, is shown in Table A9.1. Guizhou was especially hard-hit by FSZ because it was heavily dependent on taxes derived from liquor and tobacco (which had accounted for 45 percent of provincial revenues in 1993), that were reassigned as central taxes. To help cushion the shock, the MOF reportedly gave the province special consideration. Even so, provincial revenues grew faster than national revenues in 1995 and 1996 (at 24.4 percent and 27.6 percent compared to 19.6 percent and 18.0 percent for the two years, respectively), but per capita fiscal expenditure has declined in real terms. In 1995 it was nearly 25 percent less than the peak level achieved in 1991-92, making it even harder for the province to provide adequate social services to its residents. TABLE A9.1: FSZ Guizhou Province - Real Government Expenditure Per Capita, 1980-1996 (yuan) Expenditure Real Expenditure Per Capita Real Year (millions) (millions) Expenditure 1980 1209.4 1209.4 43.6 1981 1256.0 1227.3 43.4 1982 1413.9 1354.5 47.1 1983 1554.6 1479.1 51.0 1984 2144.1 1990.0 67.9 1985 2387.6 2057.8 69.3 1986 3038.7 2487.7 82.7 1987 3160.0 2410.5 79.0 1988 3814.4 2420.9 77.0 1989 4589.2 2480.7 77.9 1990 4858.1 2590.0 80.0 1991 5587.8 2883.9 88.2 1992 6063.3 2913.6 88.3 1993 6738.9 2820.9 84.7 1994 216.2* 73.7 1995 245.0* 68.8 1996 281.7* 72.5 Sources: Wong (1997) Table 4.4; Guizhou Statistical Bureau (1994), SSB 1995, 1996 and 1990 and Lou and Ge(1997). Note: Deflator used is provincial consumer price index. * per capita. 126 ANNEX 9 The Pilot Intergovernmental Transfer Scheme of 1995-1996 A transitional transfer payment scheme was introduced in 1995 as the first step toward a formula-driven redistributive system. The initial formula had two parts: an objective factor that attempted to measure the gap between "standard expenditures" and local fiscal capacity, and a policy component that directed subsidies to regions with large ethnic minority populations. In 1996 a third factor was added to the formula to reward good tax effort. Since then more tinkering has been done. The 1996 formula for transfers was as follows: Transfer to province i = f(measured fiscal shortfall of province i) + g(special transfer to province i as a minority region) + h(province i's good tax effort) The fiscal shortfall is measured as standard expenditure = standard wage expenses + standard administrative expenses + agriculture and other productive expenditures + other expenditures where standard wage expenses are derived from standard wages, number of civil servants, and a regional wage factor. standard administrative expenses are those for government administration, police and security and other government agencies. In 1995 the actual expenditures for all government units were included. In 1996 this was shifted to including personnel and running costs for fully-funded units and lumpsum costs for units that received only partial funding from the budget. agriculture and other productive expenditures are expenditures for agriculture and other "productive" departments. other expenditures include price subsidies. The policy component, or special transfers to province i as a minority region, also calculates the fiscal gap between the minority regions and the national average as (NR - PRi) x POPi where NR = national per capita revenue PRi = province i's per capita revenue POPi = population of province i The figures were 9.5 billion Yuan for quota subsidies, 36 billion Yuan for earmarked subsidies, out of a total of 54.5 billion Yuan in total central transfers (Table 1.4, Wong 1997). 127 ANNEX 9 The coefficient a, and a, are determined ex post, as the ratio of funds available for transfer divided by the size of the gap. For example, in 1996 the central government had Y2.2 billion to devote to equalization transfers, compared to a fiscal gap of Y63 billion. So a, was derived as 0.035. Similarly, the total fiscal gap for minority regions was estimated at Y13 billion in 1996, while the amount allocated to filling the gap was only Y1.2 billion in the central budget, the coefficient a, was derived as 0.09. Finally, the tax effort reward was derived in 1996 as follows: if province i had revenue growth in 1995 that exceeded the national average, then the total transfer to province i would be supplemented by the coefficient a3, where a. = 0.5 x (r' - r) and ri = revenue growth of province i r = national revenue growth so that (Transfer to province i)t = {(1+ a3i) x [ al x (standard expenditurei - PRi x POPi)+ a2 x (NR - PRi) x POPilJt-1 and t = current year, t - I = previous year In other words, the transfer to province i for 1997 will be based on its fiscal gap and tax effort in 1996. While this scheme is a welcomed first step toward a formula-based system of intergovernmental transfers, it should be recognized as a transitional scheme aimed at solving the urgent problem of meeting payroll in some provinces. It does not obviate the urgent need to get a more appropriate transfer scheme installed. Recommendations for Improving the Pilot Scheme For the transfer scheme to be effective, it must be more adequately funded. The scheme was allocated only Y2 and 3.46 billion in 1995 and 1996, respectively. This comprised just over one percent of the total of nearly Y300 billion in central transfers to the provinces in 1996 and was dwarfed by the tax rebates of Y195 billion. Because of the small sums allocated, their effect on the distribution of fiscal resources is marginal: the coefficient a, and a, are very small, so that the scheme provides only a very weak link between a province's fiscal need and their transfers. At present the scheme mixes two sets of considerations: fiscal need and support for ethnic minority regions. These objectives should be kept separate, with one equalization scheme that provides transfers according to need (plus a tax effort factor), and the other scheme to support ethnic minorities, reported separately. The total of Y3.46 billion spent on the pilot 128 ANNEX 9 scheme in 1996 was in fact split into Y2.2 billion for the equalization scheme, and Y1.2 billion for the minorities' scheme. The transitional scheme duplicates the old pre-1994 equalization transfers under quota subsidies, which totaled Y11.1 billion in 1996, and they were heavily biased toward minority regions. These schemes should be merged as soon as possible-if the pilot scheme is considered an improvement over the quota subsidies, then it should simply absorb and replace the quota subsidies. The tax effort measure takes the gap between national revenue growth and the provincial revenue growth. This is a convenient proxy that should be replaced by better measures, as better data becomes available. Provincial revenue growth depends on too many factors other than tax effort, most notably economic growth and structural change. Finally, it is urgent that a model that measures fiscal needs more appropriately replace the pilot transitional scheme. At present, the measured fiscal gap only considers the personnel and running costs of government, with wage costs weighing heavily. This reflects the short-term concern in meeting payroll requirements. In the long run, however, it is more appropriate to look at fiscal needs in terms of the costs of providing services such as education, healthcare, government administration, public transport, water, sewerage and sanitation services. This package of fiscal needs should contain not only wages and running costs but also some capital costs for building schools, clinics, roads, etc. Regional cost differentials in providing these services must also be included in the calculation of fiscal need-the present regional factor contains only a small differential for wage supplements and is insufficient. Expenditure Assignments The single biggest problem in expenditure assignments in China is the separation between spending authority and financing responsibility. On paper the intergovernmental division of expenditure responsibilities in China is roughly in conformity with that dictated by public finance theory and reflected in international best practices (see Annex 10). The central government is responsible for national defense, external relations and overall economic development, while local governments are responsible for day-to-day administration (including maintenance of public order), social services and local economic development. Some sharing of responsibilities is inevitable, especially in investments for economic development-capital construction, human development and research and development activities. However, this rational division of responsibility is often undermined by the pernicious legacy of central planning, under which the central government continues to make policies that affect spending at the lower levels, usually without compensation or even consultation. Some egregious recent examples of such practices include the policy on grain pricing and procurement, which reportedly cost the government of Jilin Province more than Y 2 billion in trading losses in 1996, or one-quarter of the provincial budgetary revenues (Fang 1997, p. 7).2 Another was the civil service wage reform in 1993, which raised pay rates for all civil servants and increased the cost of administration for central and local governments alike. Yet another example was the 2 The policy calls for government grain bureaus to procure all offered grain at the state protective price regardless of market prices. The trading losses were due to a bumper harvest in Jilin during 1996, when market prices fell. 129 ANNEX 9 decision in 1997 to create pension pooling for SOE workers at the municipal level, which created a potentially ruinous burden for municipal finances at the prevailing high stipulated pension levels, low retirement ages and low contribution rates. Many other examples can be cited where policy was made at the central level even though fiscal decentralization had transferred the burden of financing to lower levels of government-these are the source of recurrent local complaints that "the central government throws dinner parties but leaves local governments to pay the bills". The Budget Law implemented in 1994 guarantees local budget autonomy, but does not spell out the exact division of responsibility between levels of government, nor does it protect local budgets from unfunded mandates imposed by higher levels of government. The comprehensive fiscal reform package similarly neglected to redress issues of expenditure assignment even while it shifted revenues significantly to the central level. The different pace of incremental changes in revenue and expenditure trends means that after nearly two decades of market transition, intergovernmental revenue and expenditure assignments are significantly mismatched. This problem has been exacerbated since 1994: the reduced share of local revenues means an increased dependence on central transfers, especially in earmarked grants whose uses are stipulated by central policy. The spate of recent laws and regulations stipulating budgetary expenditure on agriculture to increase at a rate exceeding revenue growth, or for expenditures on education to reach 4 percent of GDP, etc., all affect recurrent expenditures in local budgets, further eroding allocative powers of local governments. Finally, the attempt to extend budgetary control over EBF will further undermine local government resources and allocative power. All of these run counter to the spirit and intent of the Budget Law. Some steps to fix intergovernmental fiscal relations: 1. Amend the Budget Law to confer real autonomy on local budgets by protecting them against imposition of unfunded mandates. Local governments should be given the right to demand compensation for costs imposed by central policies. 2. For services deemed to be of national importance, the government should consider re- centralizing their financing. In many OECD countries, for example, the central government retains some financing responsibility for health and education even though services are provided at the local levels. 3. The large income disparities across provinces in China make a compelling case for allocating substantial funds to equalization transfers to ensure minimum standards for key social services such as basic education, public health and social security. A better intergovernmental transfer scheme should be adopted that pegs transfers to measured fiscal needs. This scheme should absorb and replace the "old" quota subsidy scheme. Funds devoted to such a scheme should start off at a modest but meaningful level and increase gradually. 4. Local governments should be given sufficient revenues to finance the bulk of their expenditure responsibilities-achieving a better match between expenditure and revenue assignments to encourage better tax effort at the margin. This should be coordinated with the reform of EBF. To enlist the help of citizens in preventing the imposition of unreasonable levies, the government may consider setting up mechanisms to allow citizens to lodge complaints. 130 ANNEX 9 5. Local governments should be given greater autonomy for setting local taxes and users' fees. They should also be permitted to borrow under supervision for infrastructural investments. 6. The government should clearly define the revenue division and tax assignments for the different levels of government, ensuring the uniform extension of FSZ to subprovincial levels. Since the most important service responsibilities-for basic education, public health, and day-to-day administration -reside at the lowest levels of government, and as these levels are currently under-funded, the provincial level should not be allowed to retain additional funds. Instead, the formula used for center-provincial sharing should also be applied to sub-provincial sharing. As for the fifteen local taxes, they should all be allocated to the same level of government to avoid the expense of setting up local tax bureaus at more than one level. Thus they should be allocated to the level of government that is mainly responsible for providing health, education, and social services. 7. The system of earmarked grants should be revamped. At present there are hundreds of categories of earmarked funds at the provincial, prefectural and county levels. Most of these grant categories are legacies of the period of more centralized budget management, especially in the allocation of investment funds. Most categories are under-funded, so that earmarking is either ineffective or hampers the ability of local governments to respond to local needs. Earmarking should be confined to those few cases where investment projects have regional or national impact, or where it will ensure that aid for poor communities does reach its intended beneficiaries. 131 ANNEX 10 ANNEX 10: EXPENDITURE ASSIGNMENT: THEORY AND SOME INTERNATIONAL EVIDENCE1 Introduction Government involvement in the economy differs widely among countries, both developing and developed. General government expenditure and net lending in 1987 was on average 41.3 percent of GDP for the industrialized countries and 35.18 percent of GDP for developing countries, but it was as low as 9.7 percent for Paraguay and as high as 55 percent for Brazil and 64 percent for Hungary in the same year.2 For China, the figure is about 21.6 percent of GDP for 1994. Perhaps less quantifiable but as pervasive and diverse as expenditures is government involvement in the economy through regulation and ownership of assets. The variety of government involvement reflects the diversity in norms and values between societies, of which government is an agent. As China becomes more and more akin to a mixed economy, government should concentrate on those goods and services for which it has a comparative advantage in provision, a topic dealt with in the theory of public goods.3 Whether government has a comparative advantage over the market depends, among others, on the organization of government, and the division of tasks over levels of government, the topic of the first part of this Annex. This division of tasks, or expenditure assignment is logically prior to revenue assignment,4 but is not independent, because the incentives for lower levels of government to perform tasks assigned to them depend on the way those tasks are financed. Revenue assignment will be treated in the second part of this annex. This is adapted from Annex 3.1, World Bank, China: Budgetary Policy & Intergovernmental Relations. Report No. 11094, CHA, October 30, 1992. 2 IMF, Government Finance Statistics Yearbook, 1990. See, for example, R. Musgrave and P. Musgrave, Public Finance in Theory and Practice, 4th ed., McGraw- Hill, New York e.o., 1984. See Anwar Shah, 1991, "Perspectives on the Design of Intergovernmental Fiscal Relations," World Bank. WPS 726, p. 2. For a discussion of tax assignment principles, see R.A. Musgrave, "Who Should Tax, Where and What?" in Charles E. McLure, Tax Assignment in Federal Countries, Australian University Press, Canberra, 1983. 132 ANNEX 10 Criteria for Expenditure Assignment over Levels of Government Many goods and services provided by government have benefits restricted to only a part of a country, for example waste collection, local roads and police. From a normative point of view, it does not matter what level of government provides such a local public good, but a positive perspective on government-government as a complex organization of people with different interests-leads to the conclusion that decentralization, if properly implemented, can enhance economic efficiency within government and promote growth due to better policies at a subnational level. China's economic growth following the decentralization over the reform period is a prime example of this. Decentralization is concerned with at least five aspects of a function: (a) the choice of objectives; (b) the choice of instruments; (c) the administration and implementation of a program; (d) the financing of the program; and (e) the control of a program. Usually one distinguishes three types of decentralization, depending on the degree of autonomy of subnational governments and depending on whether decentralization takes place within the administration: (a) deconcentration: subnational units of central government administer and implement national policies; (b) delegation: subnational governments perform policies decided upon by the national government, but with some autonomy in the choice of instruments or finance is present; (c) devolution: subnational governments decide on goals, instruments, implementation and finance of a policy, but sometimes with cofinancing and central government control. In China, due to the double accountability of subnational government units to both local People's Congresses and central government units, the distinction between the types of decentralization is not always clear-cut. For instance, local Education Bureaus act as agents of the Ministry of Education concerning national universities (deconcentration); they have some autonomy concerning finance and instruments of primary education, if national goals are met (delegation); and they are largely autonomous concerning local universities (devolution). Many problems between levels of governments stem from partial decentralization of responsibility. An objective imposed by the central government is least likely to be fulfilled by a local government that disagrees, and it is less likely to perform tasks for which it is not explicitly or implicitly held responsible. Prescription of instruments for local governments to achieve certain goals may lead to inefficient implementation of the program, if local circumstances are not taken into account. Decentralization of programs without the means of finance to go with it has implications for the design of intergovernmental grants, or may put the subnational government under undue budgetary stress, if no additional finance is provided. The theory of fiscal federalism' provides a strong rationale for the decentralization of decisions concerning local public goods, concisely stated in Oates' decentralization theorem.6 "For a public good-the consumption of which is defined over geographical subsets of the total population, and for which the costs of providing each level of output of the good in each jurisdiction are the same for the central or the respective local government-it will always be more efficient (or at least as efficient) for local governments to provide the Pareto-efficient levels of output for their respective Fiscal federalism applies to any government with more than one layer of government, irrespective of whether the country is a federation or not. Wallis E. Oates, 1972, Fiscal Federalism, Harcourt Barce Janovitch Inc., New York e.o., p.35. 133 ANNEX 10 jurisdictions than for the central government to provide any specified and uniform level of output across all jurisdictions." The reasons most cited in favor of decentralization are: efficiency, accountability and the benefits from competition among subnational governments. (a) Efficiency. The level of government closest to the beneficiaries of the public good should make the decision on provision of public goods. The information on local preferences for public goods and on local costs is better, resulting in higher allocative and production efficiency. Also, better information on local circumstances enhances the efficiency of public investment decisions, and fosters growth. (b) Accountability. Local governments can be held more directly responsible for their decisions than national ones. Besides sanctions through the political process, "voting with one's feet" can sanction local decisions by leaving the jurisdiction, which erodes the local tax base, in principle forcing the government to adjust. The corrective mechanism of exit only works if the tax base is allowed to change in reaction to bad policy and if interest of the local government in the local tax base exists. If local governments were financed completely by central government grants, they would have no interest in conducting policy to the benefit of their constituents, as their sources of revenue does not depend upon it. Moreover, transparency of government decisions and their consequences is a condition for the "exit" and "voice" mechanisms to work. Besides political accountability, financial accountability and managerial accountability can be distinguished. Financial accountability refers to government spending being in accordance with the appropriations and is usually exercised by parliamentary committees or auditors from higher-level government. Managerial accountability refers to the efficiency of public goods provision, if a lower level of government acts as an agent of the central government. The less political accountability is, the more important financial and managerial accountability becomes. Decentralization improves accountability since it reveals information about the most efficient way of providing a certain service. Thus, even if the objectives and responsibility of a program lies with central government, subnational governments may outperform a national agency in administration and implementation of the policy. For financial and managerial accountability to be effective, an accurate accounting system, with regular and detailed reviews of government operations in financial and economic terms, is a requirement. (c) Competition and Innovation. As in markets for private goods, competition among jurisdictions may enhance the provision of efficient public goods, because new and more cost-efficient methods are more likely to be found when more jurisdictions search independently for them. The best practice thus found can be imitated by other localities. Local government's interest in its own tax base is again essential for the competitive mechanism to work. Of course, learning from others is not restricted to national boundaries, but national innovations may be more relevant to national circumstances. A more subtle gain from competition arises, once imperfect information is considered. In such circumstances, mistakes are due to occur, but the impact of the mistake is far less pervasive in a decentralized system than in a centralized system in which an erroneous policy is implemented nationwide. 1 134 ANNEX 10 Not all government tasks can be decentralized. Arguments against the decentralization of government functions are: (a) Imperfect Mobility of People. The corrective mechanism of voting with one's feet-leaving a jurisdiction with a suboptimal level of public goods-may be inoperative if the costs of moving are high. Legal restrictions against the mobility of people equally give the local government more discretion in diverging from the preference of its constituents. In various countries, the political process is inadequate for the expression of local preferences concerning local public goods. Unfortunately, the same situation usually prevails at higher levels of government, and thus centralizing the provision of local public goods may not improve the situation in this respect. However, if the mechanism of voting-both with one's feet and at the ballot-is not in operation, managerial accountability of local governments may partially replace it. In such a situation, efficiency would still be enhanced by decentralization, because then the central government has more evidence to judge the individual local government on the efficiency of providing a service. In addition, the competitive argument mentioned above also holds if local governments only face managerial accountability to higher levels of government. (b) Spatial Externalities. Local public goods may have positive or negative effects in other jurisdictions, which are not taken into account by the government supplying them. The limiting case is a national public good, such as national defense, or foreign policy, which ought to be supplied by the central government. More often, the spillovers are limited to neighboring jurisdictions (people from a metropolitan area using facilities in a city, such as roads, museums), and could in principle be dealt with by horizontal intergovernmental financial compensation, or other forms of intergovernmental cooperation, without centralizing the provision of the service. Broader application of user charges for services with spillovers may reduce the importance of spatial externalities, because then the benefits of services would be internalized. Local government competition may lead to spatial externalities: local governments can offer benefits to their constituents at the cost of others outside the constituency. Examples are: the loosening of environmental standards in a locality, when the environmental degradation takes place in another locality due to spillovers; the imposition of high taxes on a local firm, which has a national monopoly and can thus shift the tax burden to nonconstituents; and the exemption of certain taxes of which the revenue is only partially local. As with externalities in the private sector, detrimental effects from competing local governments are often due to an inappropriate regulatory framework, in which the full costs and benefits do not accrue to the localities themselves. The fear of tax competition among jurisdictions sometimes keeps governments from decentralizing the tax base, thereby reducing the responsibility of the local government for its level of services. Tax competition occurs when local governments set low rates of taxation for mobile means of production, especially capital. Without coordination among jurisdictions, the outcome may be an inefficiently low national level of taxation, and therefore a suboptimal supply of public goods. The same argument would hold for other governmental intervention, such as regulation, trade policy and industrial policy, where individual policies by local governments may add up to inefficient national outcomes. Tax competition may also affect the efficiency of the private sector, because costly reallocation 135 ANNEX 10 of resources may be undertaken solely for tax purposes. To contain the detrimental effects of tax competition, the central government can set minimum standards of public services and taxes, or may even centralize certain functions. Functions determining the internal market of a country, such as banking and currency, trade policy and industrial policy are normally central functions in order to prevent destructive competition among local governments. The conduct of local stabilization policy is in general not effective, due to the large spillovers across local boundaries. Increased government spending in one province will "leak" over the borders of another, resulting in too little concern with a stable economic environment. Even if no active stabilization policy by local governments is pursued, however, decisions on expenditures and taxes do affect the overall economy, and the decentralization of decision power might compound the problems for the central government in order to contain, or compensate for, those effects. Perhaps the most pervasive cyclical influence stems from government borrowing, and some means to control the impact of local borrowing will be discussed in the section on revenue assignment. (c) Economies of Scale. Some services provided by governments need a minimum scale in order to operate efficiently, a scale that may not be achieved by a single locality. Again, economies of scale do not automatically pall for higher-level government intervention: intergovernmental cooperation may often be more efficient, as local governments can still be held responsible for the provision of the service. Electricity nets, sewage systems, waste collection, etc. are services that might be shared by several local governments in cooperation, thus bypassing the negative effects of administration by a higher-level government. (d) Administrative and Compliance Costs. It is argued that a centralized administration leads to lower administrative and compliance costs for certain programs. For instance, redistributive tax policies are best carried out by central government, since decentralized policies to redistribute income would "drive out the rich". However, although it is clear that an individual cannot escape its tax liability from a national tax by moving, it may well be that local governments have more capabilities to monitor actual tax obligations. Consequently, the administration and implementation of a tax policy with national objectives may be delegated to subnational governments, and a share of the tax may be left at the collecting agency in order to align the goals of central and subnational governments. On the expenditure side, subnational governments may be in a better position to assess the needs for welfare support, and could take more interest in providing the right amount of support to the right individuals, if part of the welfare payments is locally funded. A more serious hindrance for decentralization in many developing countries is insufficient administrative capacity at lower levels of government, leading to production inefficiencies. (e) Equity. Local income redistribution is constrained by mobility of people: a locality that has strong redistributive policies would drive out the rich and attract the poor, thereby eroding its own tax base. Even if local redistribution were feasible, the inequalities between localities would not be affected. Both reasons give strong motivation for nationwide redistributive policies. The desire to supply every citizen with a minimum amount of certain local public goods and publicly provided private goods (education, health care) is brought up as an argument 136 ANNEX 10 against decentralization. The argument holds for deciding the objectives of a program, but administration, implementation strategy and financing can often be left to the discretion of lower levels of government. The problem is rather to find accurate ways to decentralize most of the decisions in order to reap allocative and productive efficiency gains, while finding ways to impose the minimum standards on the lower-level governments. National standards, central inspection agencies and appropriate grant design are often more efficient than the complete centralization of provision. Private producers or local governments can ensure personal entitlement to education or health care in the form of education vouchers and social health insurance, while maintaining the competition on the supply side of the public good. Equity issues also play a role in the assignment of revenue sources. The decentralization of revenue sources may lead to horizontal inequities to different treatment of people in the same material circumstances, but in different localities. The equity issue can only be dealt with in the context of net benefits from the public sector: individuals may well be willing to pay more taxes for better services. Furthermore, the issue becomes policy-relevant only if there are serious impediments to mobility: if people are free to move, they settle in the locality that offers the best deal to them. Even if some inequity is created by decentralization, it should always be weighted against the efficiency gains from decentralization. Based on the theoretical arguments, a tentative and broad division of labor over levels of government can be proposed (Table A10.1), distinguishing between service responsibility and service provision. The first is concerned with the objectives of a program or service, the second with the actual delivery of the service, and the two need not coincide. Table AlO.2 shows how governments around the world assign functions among levels of government. Table A10.3 shows the actual expenditures by levels of government for a sample of 39 developing and developed countries over a three-year period. The central government administers most expenditures: in all but four countries listed, it accounted for over 50 percent and an average share in total expenditures of 72.3 percent. The spectrum is wide, ranging from Yugoslavia (23 percent) to Paraguay (95 percent). Central governments have a larger portion of expenditures in developing countries (77.8 percent) than in developed countries (65.9 percent). Moreover, service provision need not coincide with service production. For instance, local governments may be assigned the provision of local infrastructure, but the actual production of infrastructure may very well be left to enterprises outside the public sector. 137 ANNEX 10 Table A10.1: Conceptual Basis of Expenditure Assignment Service Service Comments Expenditure category responsibility provision Defense F F National benefits Foreign affairs F F International trade F F Environment F S,L Currency, banking F F Interstate commerce F F Immigration F F Unemployment insurance F F Airlines/railways F F Industry/agriculture F,S,L S,L Interstate spillovers Education F,S,L S,L Transfers in kind Health F,S,L S,L Social welfare F,S,L S,L Police S,L S,L Primarily local benefits Highways F,S,L S,L Some roads with spillovers; others primarily local Natural resources F,S,L S,L Promotes a common market F=central government; S=intermediate government (province, state); L=local government. Source: Anwar Shah, "Perspectives on the Design of Intergovernmental Fiscal Relations," PRE Working Paper Series, No. 726, Washington, D.C., July 1991, p. 6 and p. 7. 138 ANNEX 10 TABLE A10.2: EXPENDITURE ASSIGNMENT IN SELECTED COUNTRIES Foreign Int'l Environ- Cur. Interstate Unempl. Air Industry Educa- Social High- Natural Residual Defense Affairs Trade ment Bank Trade Immig. Insur. Rail Agriculture tion Health Welfare Police ways Resources Powers Argentina: (1) F F F F F F,S F.S.L F,S,L F,S F,S F,S (2) F F F F F F,S F.S.L F,S,L F.S F:S F,S S Bangladesh: (1) F F F F L L L L L L L (2) F F F F L L L L L L L Brazil: (1) F F F F,S F F F F F F.S F F,S F,S F,S F F,S S (2) F F F F.S F F F F F F,S F,S,L F,S,L F,S F, S F,S FIS,L S Czechoslovakia: (1) F F F F F F S L L S F (2) F F F F F F S L L S,L F India: (1) F F F F,S F F F F,S F F,S F,S S F,S S F F,S F (2) F F F F,S F F F F,S F F,S F,S,L S,L F,S S F F,S F Indonesia: (1) F F F F F,S,L F,S,L F,S,L (2) F F F F F,S,L F,S,L F,S,L Japan: (1) F F F L F,L F,L F,L L L (2) F F F L F,L F,L F,L L L Malaysia: (1) F F F L F F F F F,S F F,S F,S F F S F (2) F F F L F F F F F,S F F,S F,S F F S F Mexico: (1) F F F F F F F F F F,S F,S S,L S (2) F F F F F F F F F F,S F,S S,L S Nigeria: (1) F F F F,S F F F F F,S F,S S,L F F,S,L F,S,L F (2) Pakistan: (1) F F F F F,S F F,S F,S,L F,S,L F,S S F,S F S (2) F F F F F F,S S,L S,L F,S S F,S F S Philippines: (1) F F F F F F F (2) F F F F F S,L S,L S,L Soviet Union:/a (1) F F F F F F (2) F F F F F F,S,L F,S,L F,S,L F,S,L F Thailand: (1) F F F F L L L L (2) F F F F L L L L Yugoslavia: (1) F F F F F F F S S S F F (2) F F F F F F S,L S,L S,L (assignment of constitutional powers) Canada F F F F,S F,S F C F F,S C S S(F) F,S F,S S F,S F United States F,S F F F,S C F C F,S F,S S S,F S(F) F,S F,S S S Switzerland F F F C F F C C F,C F,S C,F,S S,C F,C S FS S Australia F,S F C F,S C C C C F,S S,C F,S F,S C S,F F,S F S Germany F F,S C C C C F,C C C,S C,F,S C C,S C C S Austria F F F,S F F F F F F,S C,F,S C F,S F,S C S /a Allocation of Expenditure responsibility between republics and local governments is still an open question. For row (2), the table reflects actual distribution of main expenditures between different budgets in 1989. Notes: F = Federal/Central Government; S = State/Provincial/Republic/Departmental Government; L = Local Government; C= Concurrent Powers; (1) = Service Responsibility; (2) = Provision of Service. Sources: Anwar Shah, Perspectives on the Design ofIntergovernmental Fiscal Relation; Policy Research Working Paper Series No. 726, July 1991; DHerpinger, Distribution ofPowers and Functions in Federal systems, Ottawa, 1991, Appendix B. 139 ANNEX 10 TABLE 10.3: GENERAL GOVERNMENT EXPENDITURES AND PORTION ADMINISTERED BY EACH LEVEL OF GOVERNMENT, SELECTED COUNTRIES /a (Average of latest three years available) Social Social security Total security Total Expenditures /c Education Health and welfare End expen- Educa- and Central State Local CentgI State Local Central State Local Central State Local year diture tion Health welfare gov't gov't gov't govW gov't gov't gov't gov't gov't gov't gov't gov't ..-...-.-.. (as percent of GDP) ------------ ------------------------------------------------- (as percent of general government) --------------------------- Australia 1987 39.1 5.5 5.5 9.6 52.9 40.4 6.8 8.5 91.3 0.2 43.5 55.6 0.9 92.8 6.2 1.0 Austria 1987 51.8 .. .. .. 70.4 13.7 16.9 .. .. .. .. .. .. Belgium 1987 56.7 .. .. .. 85.9 .. 11.9 .. .. .. .. .. .. Canada 1987 46.0 5.8 6.0 12.3 41.3 40.3 18.4 4.8 34.5 60.7 2.6 89.5 7.9 65.8 31.3 2.9 Denmark 1986 57.6 7.1 5.2 23.1 44.9 .. 52.9 46.8 .. 53.2 7.1 .. 92.9 26.1 .. 73.9 Finland 1987 43.0 .. .. .. 54.7 .. 45.3 .. .. .. .. .. .. .. France 1985 49.3 4.6 8.3 20.9 82.2 .. 16.5 75.3 .. 24.7 97.0 .. 3.0 91.8 .. 8.2 Germany 1983 50.2 4.2 8.0 21.2 58.7 21.5 17.9 1.0 73.8 25.2 74.4 11.2 14.4 79.0 10.9 10.1 Ireland 1987 55.8 .. .. .. 72.5 .. 27.5 .. .. .. .. .. .. Luxembourg 1987 39.1 4.4 0.7 21.3 81.3 .. 25.9 . 74.1 .. 25.9 92.0 .. 8.0 97.4 .. 2.6 Netherlands 1988 59.2 .. .. .. 70.1 .. 29.9 .. .. .. .. .. .. New Zealand 1981 43.2 .. .. .. 86.9 .. 13.1 .. .. .. .. .. .. Norway 1986 47.2 .. .. .. 66.4 .. 33.6 .. .. .. .. .. .. Spain 1986 38.2 .. .. .. 78.7 9.9 11.3 .. .. .. .. .. .. Sweden 1987 61.6 .. .. .. 59.8 .. 40.2 .. .. .. .. .. .. Switzerland 1984 37.4 5.3 5.9 13.9 47.5 28.3 24.2 6.2 57.5 36.3 45.5 32.1 22.4 88.5 5.6 5.9 United Kingdom 1987 44.8 5.1 5.1 14.3 70.9 .. 27.2 12.7 .. 87.3 100.0 .. 0.0 84.0 .. 16.0 United States 1987 37.1 5.1 4.3 9.0 60.3 17.3 22.4 4.2 24.5 71.3 50.5 33.8 15.7 78.0 14.6 7.4 Hungary 1988 64.5 5.7 4.2 18.1 77.8 .. 22.2 20.2 .. 80.0 39.2 .. 60.8 95.7 .. 4.3 Poland 1988 48.1 .. .. .. 77.1 .. 28.9 .. .. .. .. .. .. Romania 1985 32.3 2.1 2.1 8.9 77.0 .. 23.0 28.0 .. 72.0 10.3 .. 89.7 99.3 .. 0.7 Yugoslavia 1987 25.3 3.2 4.2 7.8 23.2 31.4 45.4 0.0 0.0 100.0 0.0 0.0 100.0 7.3 75.9 16.8 India /b 1986 22.6 3.4 0.9 2.3 47.5 52.5 .. 9.0 90.1 .. 30.2 69.8 .. 0.0 100.0 Indonesia lb 1988 22.8 3.1 0.5 0.4 88.7 11.3 .. 65.3 34.7 .. 72.8 27.2 .. 0.0 100.0 Israel 1986 62.9 5.3 2.0 10.0 90.8 .. 9.2 67.2 .. 32.8 97.0 .. 3.0 94.9 .. 5.1 Pakistan 1979 26.1 .. .. .. 68.2 28.3 3.5 .. .. .. .. .. .. Thailand 1982 21.2 4.1 1.1 1.2 92.3 .. 7.7 94.8 .. 5.2 93.5 .. 6.5 97.4 .. 2.6 Argentina /b 1987 33.2 4.0 1.1 9.1 60.3 39.7 .. 33.3 66.7 .. 24.4 75.6 .. 89.4 10.6 .. Bolivia 1986 11.1 .. .. .. 85.9 10.6 3.4 .. .. .. .. .. .. Brazil 1987 34.1 .. .. .. 65.8 24.5 9.6 .. .. .. .. .. .. Chile 1987 32.3 4.9 1.9 8.8 93.8 .. 6.2 81.7 .. 18.3 98.1 .. 1.9 100.0 Colombia 1984 18.0 5.5 1.3 3.2 67.4 23.9 8.7 55.5 39.2 5.3 49.0 40.2 10.8 90.0 7.8 2.2 Mexico 1984 30.2 .. .. .. 90.1 7.6 2.3 .. .. .. .. .. .. Paraguay 1984 11.3 .. .. .. 95.1 .. 4.9 .. .. .. .. .. .. Kenya 1984 29.3 5.2 2.1 1.4 94.3 .. 5.7 94.0 .. 6.0 91.9 .. 8.1 75.9 .. 24.1 Malawi 1984 29.1 3.7 2.2 0.6 93.7 .. 6.3 98.7 .. 1.3 82.9 .. 17.1 100.0 .. 0.0 South Africa 1986 33.3 .. .. .. 74.8 12.5 12.7 .. .. .. .. .. .. Tunisia 1982 34.0 5.1 2.5 4.7 94.6 .. 5.4 100.0 .. .. 100.0 .. .. 100.0 .. 0.0 Zimbabwe 1986 45.0 8.3 2.6 3.0 75.8 .. 24.2 60.2 .. 39.8 .. .. .. 100.0 .. 0.0 USSR /d 1990 50.6 5.2 2.9 7.4 RussianFederation /d 1992 41.0 .. /a Excluding intergovernmental grants. 75 Data for general government do not include local government. 7c Includes supranational authorities share of general government expenditures. Source: Levin (1991). 7a Data for USSR and Russian Federation is from Table 1.3 of World Bank Report "Russian Intergovernmental Fiscal Relations in the Russian Federation." Note: General Government is central plus subnational government plus social security agencies. 140 ANNEX 11 ANNEX 11: A FORMULA-BASED EQUALIZATION TRANSFER SYSTEM FOR CHINA: MODEL AND SIMULATIONS The current Chinese central-provincial fiscal transfer system consists mainly of three mechanisms! The first mechanism is based on the old contract system prevailing during 1988-1993. That is, after 1994, the localities (provinces and cities with independent planning status) continue to remit revenues to or receive transfers from the center according to their fiscal contracts in effect in 1993. The second type of transfer is "returned revenue" from the center according to a calculation that ensures that each locality retains no less than what it did in 1993. These two types of transfers are general-purpose transfers, or unconditional transfers. The third type of transfer includes various specific-purpose grants, such as those for price subsidies, educational projects, environmental projects, disaster relief and the development of poor regions. The general-purpose transfers from the center to the localities account for the major part of the total transfers. In 1994, the central government's net transfer to the local governments amounted to about 181 billion Yuan, of which two-thirds were general-purpose grants. However, these general-purpose transfers suffer from at least two major flaws. First, they were not designed to address the increasingly significant regional disparity issue; rather, they were largely designed to recognize the vested interests of the localities? Second, the criteria by which these transfers are allocated are rather ad hoc; that is, the transfer system lacks scientific measurements of fiscal capacities and fiscal needs. This can easily lead to an unjustified distribution and encourage the localities' bargaining activities. This annex presents an illustrative equalization transfer model and the simulation result using 1994 data on China. It provides the estimates of fiscal capacities and fiscal needs of 30 Chinese provinces using the methods discussed in the paper. Based on a formula that aims to ensure that provinces with similar levels of tax effort be able to provide similar levels of public services, the calculation results in a set of hypothetical transfers from the center to the provinces in 1994. These results are then compared with the actual transfers made to the provinces in 1994. The method used to calculate the provincial fiscal capacities and fiscal needs may be considered overly simplified and the quality of data can certainly be improved. Nevertheless, the estimation carried out here is simply intended to provide an illustrative example of how an equalization transfer formula with a minimum data requirement can be constructed. The following sections discuss the methodology and the results. Estimating Fiscal Capacities In 1996, a formula-based equalization transfer system was introduced on an experimental basis. This system applies a formula that uses objective variables to calculate local fiscal capacities and needs. However, the size of this transfer program was only $2 billion Yuan, or 0.5 percent of the central government revenue. 2According to some studies, China's regional disparity is among the highest in the world. 141 ANNEX ]1 Two methods were tried to determine the fiscal capacities of the provinces, that is, the revenue each province would be able to collect with an average level of tax effort. The most important element in this calculation is the estimation of the provinces' tax bases. The first method uses provinces' GDP levels as proxies of the tax bases. To see how well GDP can forecast revenue, an OLS regression of a linear equation with no intercept is conducted. The result is: Regression I: REVi = 0.0186GDPi (1) R-square = 0.75, No. of observations = 30, Degrees of freedom =28 where REVi represents revenue collected province i in 1994, and GDPi is the value of province i's gross domestic product in 1994. The result suggests that the differences in GDP can explain 75 percent of the variations in revenue across provinces.3 Instead of using an output measure such as GDP as a proxy of the tax base, the second method attempts to estimate the local tax base using two variables: the total retail sales and the before-tax profits of industrial enterprises. This is based on the assumption that business tax and personal income tax-the two major local taxes-are positively correlated with total retail sales, and another major local tax-corporate income tax-is positively correlated with profits of industrial enterprises4 Using these two variables as explanatory variables, the second regression yields a better fit: Regression II: REVi = 0.0896 SALESi + 0.1679 PROFi (2) (4.90) (2.84) R-square = 0.88, No. of observations = 30, Degrees of freedom =28 where SALESi is province i's total retail sales in 1994, and PROFi is the profits of the province's state- owned industrial enterprises in 1994. Roughly speaking, the regression results suggest that if all local taxes are levied on these two bases, the national average effective tax rate on total sales is 8.96 percent, and the national average effective tax rate on corporate profits is 16.8 percent. The R-square of 0.88 suggests that 88 percent of the variations in revenue collection can be explained by variations in these two variables' Because the second method represents a better forecast of actual revenues, it is used to estimate the provinces' fiscal capacities. That is, the following equation is employed to estimate the fiscal capacities of the provinces: Ci = 0.0896 SALESi + 0.1679 PROFi (3) When the structure of the economy (percentage of agriculture in GDP) is added to the regression as the second explanatory variable, R-square improves only slightly to 0.77, and its coefficient is not statistically significant. Twenty-five percent of VAT assigned to provinces is not considered as local tax in this exercise, as it is included in the transfer calculated from the actual local revenue and expenditure figures. To find out whether individual disposable income (the tax base for personal income tax) is important in determining revenue, this variable is added to the regression but yields no improvement in the R-square. When tax effort (revenue/GDP ratio) is added to the above regression, the R-square further increases to 0.94. However, to forecast the provinces' fiscal capacities, one should eliminate the impact of tax efforts by leaving out this variable. 142 ANNEX ]] where Ci is the fiscal capacity of province i. The estimated fiscal capacities are reported in Table A 1.1 Estimating Fiscal Need The fiscal need of each province is broken down into seven categories: education, health, social welfare, police and law enforcement, infrastructure, government administration and other services. For each category, a formula was constructed to determine the expenditure needs of the province. The variables used in these formulas are the most important determinants of the expenditure and are those for which data are readily available. The variables used to determine the needs under the seven categories are: Education: population, average number of years of education Health: population, average life expectancy Social welfare: population, percentage of population over age 65, urban unemployment rate Police and law enforcement: population, percentage of urban population Infrastructure: length of roads, area Government administration: population Other services: population Determining the fiscal need of each province involves three steps: Step 1: determining the share of each spending category in total spending. The share of each expenditure category in total expenditure is calculated using the actual spending figures in 1994 (Table Al 1.2): 143 ANNEX 1] Table All.l: Fiscal Revenues, Expenditures, and Estimated Fiscal Capacities Actual Tax Actual Actual Estimated revenue effort expndt. transfer fiscal (Mil. Y) (Rev/GDP) (Mil. Y) (Mil. Y) capacity 1994 1994 1994 1994 1994 All China 231159 5.1 392962 11803 228596 Beijing 4585 4.2 9853 5268 9055 Tianjin 5015 6.9 7232 2217 4563 Hebei 9522 4.4 16084 6562 9396 Shanxi 5382 6.3 8923 3541 4562 Inner Mongolia 3630 5.3 9282 5652 3188 Liaoning 15367 5.9 22358 6991 12657 Jilin 5127 5.3 10459 5332 5213 Heilongjiang 8466 5.2 14240 5774 9368 Shanghai 16962 8.6 19084 2122 14323 Jiangsu 13662 3.4 20017 6355 17909 Zhejiang 9463 3.5 15303 5840 13422 Anhui 5468 3.7 9327 3859 6263 Fujian 9194 5.5 13773 4579 6868 Jiangxi 4929 4.8 9203 4274 4175 Shandong 13466 3.5 21877 8411 16734 Henan 9335 4.2 16962 7627 9989 Hubei 7746 4.1 13720 5974 9679 Hunan 8589 5.1 15149 6560 8103 Guangdong 29870 7.0 41683 11813 24621 Guangxi 6226 5.0 12493 6267 5423 Hainan 2753 8.3 4001 1248 1008 Sichuan 13599 4.9 23739 10140 12181 Guizhou 3124 6.0 7423 4299 2325 Yunnan 7670 7.9 20373 12703 7631 Tibet 554 12.1 3030 2476 196 Shaanxi 4259 5.0 8552 4293 3670 Gansu 2908 6.4 7238 4330 2536 Qinghai 701 5.1 2536 1835 583 Ningxia 717 5.4 1938 1221 603 Xinjiang 2870 4.3 7110 4240 2352 Source: Tables Al1.1 (a) and (b), and authors own calculations. 144 ANNEX 11 Table A11.1(a): China: Basic Indicators by Province Population GDP Area Aged Urban per capita population unemployment 1994 1994 (%) rate (Mil.) (Th. Y) (Th. Mu) 1987 1994 All China 1198.50 3.76 9590.09 6.23 2.8 Beijing 11.25 9.64 16.80 7.66 0.4 Tianjin 9.35 7.76 11.31 7.79 1.2 Hebei 63.88 3.36 187.95 6.34 2.8 Shanxi 30.45 2.80 156.30 5.98 1.2 Inner Mongolia 22.60 3.02 1192.04 4.26 3.7 Liaoning 40.67 6.35 146.15 6.54 2.5 Jilin 25.74 3.76 186.81 5.56 1.8 Heilongjiang 36.72 4.41 451.47 4.17 2.5 Shanghai 13.56 14.54 6.30 11.52 2.8 Jiangsu 70.21 5.78 102.52 8.23 2.1 Zhejiang 42.94 6.21 101.77 8.20 3.1 Anhui 59.55 2.50 139.06 6.35 3.1 Fujian 31.83 5.29 121.12 6.17 2.4 Jiangxi 40.15 2.57 166.86 5.39 2 Shandong 86.71 4.47 156.57 6.72 3.1 Henan 90.27 2.44 167.01 6.33 2.3 Hubei 57.19 3.29 186.10 6.14 3 Hunan 63.55 2.67 212.10 6.43 3.8 Guangdong 66.89 6.34 177.99 7.28 2.4 Guangxi 44.93 2.76 237.34 6.20 3.6 Hainan 7.11 4.66 33.98 6.36 3.6 Sichuan 112.14 2.48 570.31 6.79 3.8 Guizhou 34.58 1.51 176.04 4.95 5.5 Yunnan 39.39 2.47 393.33 5.70 2.7 Tibet 2.36 1.94 1220.01 6.67 5 Shaanxi 34.81 2.43 205.52 5.26 3.5 Gansu 23.78 1.90 456.55 4.62 5.3 Qinghai 4.74 2.92 742.82 3.93 6 Ningxia 5.04 2.66 51.73 3.69 5.3 Xinjiang 16.32 4.13 1683.98 4.62 3.2 Source: State Statistical Bureau (1995). Cost data are from Wu Renhong, 1995, "China's Inflation and Regional Disparity," and World Bank, 1994, China: Internal Market and Regulations. Social indicators are from Yasuko Hayase and Seiko Kawamata, 1991, Population Policy and Vital Statistics in China, Institute of Developing Economies, Tokyo, Japan. Author's own calculation. 145 ANNEX ]] Table A11.1(b): China: Basic Indicators by Province Urban Length Average Average Wage& Population of roads years of life cost (%) (kms) education expectancy index 1994 1994 1987 1987 1992 All China 23.6 1117821 5.68 70.59 1 Beijing 66.6 11532 8.12 74.93 1.19 Tianjin 57.7 4156 7.42 73.64 1.02 Hebei 16.5 50496 5.90 73.26 0.98 Shanxi 25.8 32693 6.29 69.77 0.81 Inner Mongolia 37.0 44202 6.19 69.24 0.82 Liaoning 45.2 42763 7.00 73.32 0.93 Jilin 43.9 29581 6.69 70.11 0.87 Heilongjiang 49.7 48356 6.55 70.33 0.82 Shanghai 70.5 3721 7.92 75.97 1.29 Jiangsu 23.9 25891 5.91 73.63 1.04 Zhejiang 16.5 33170 5.82 71.82 1.21 Anhui 17.2 30876 4.71 71.21 0.87 Fujian 18.3 44608 5.12 70.90 1.19 Jiangxi 20.9 34556 5.15 68.10 0.86 Shandong 17.4 50225 5.52 72.88 0.90 Henan 14.8 47704 5.43 71.68 0.83 Hubei 29.0 48349 5.92 68.91 0.94 Hunan 16.5 58803 6.00 68.82 1.01 Guangdong 16.1 75716 5.96 73.83 1.57 Guangxi 14.1 39550 5.54 70.81 1.07 Hainan 34.1 13015 5.84 69.76 1.22 Sichuan 16.3 100002 5.40 68.70 0.90 Guizhou 15.4 32398 4.37 70.12 0.95 Yunnan 15.9 65578 4.13 64.25 0.94 Tibet 14.1 21842 1.91 63.50 1.17 Shaanxi 21.3 39058 6.29 69.74 0.91 Gansu 18.8 34984 4.40 70.24 0.97 Qinghai 32.6 17061 3.79 66.40 0.90 Ningxia 28.8 8324 5.15 69.74 0.93 Xinjiang 47.7 28611 6.14 69.25 1.00 Source: State Statistical Bureau (1995). Cost data are from Wu Renhong, 1995, "China's Inflation and Regional Disparity," and World Bank, 1994, China: Internal Market and Regulations. Social indicators are from Yasuko Hayase and Seiko Kawamata, 1991, Population Policy and Vital Statistics in China, Institute of Developing Economies, Tokyo, Japan. Author's own calculation. 146 ANNEX 11 Table A11.2: Local Expenditures by Category, 1994 (100 Mil Y) 1994 Actual Amount Share a Education, Culture and Related Expenses 83858 0.276 Health, Family Planning and Sports 27467 0.090 Social Welfare 9416 0.031 Government Administration 66539 0.219 Police and Law Enforcement 22000 0.072 Infrastructure Maintenance 23416 0.077 Other services (including subsidies) 71211 0.234 Subtotal 303907 1 Note: Capital expenditures, except for urban maintenance, are excluded. Infrastructure maintenance is named "urban maintenance" in China Almanac of Finance 1995. Data Source: China Almanac of Finance 1995. The total fiscal need of 30 provinces in category k (k = education, health, etc.) is the weight (ak) multiplied by the total fiscal need of all categories. Denoting total local need of all categories by TN, the total fiscal need in category k is TNk = ak*TN Step 2: determining the fiscal need of each province in category k. For education (k=E), the fiscal need of province i is calculated using the following formula: NIE= TNE(PiEi/FjPjEj) = cxE*TN(PiEi/ZjPjEj) (4) where NiE is province i's fiscal need for education, aE=0.276 is the weight assigned to education, TNE is the 30 provinces' total fiscal need for education, P, is the population of province i, Ei is the ratio of the national average number of years of education to that in province i, and PiEi/ZjPjE is the share of province i's fiscal need in the 30 provinces' total need for education. For health (k=H), the fiscal need of province i is calculated using the following formula: NiH = TNH(PiLi/FjPjLj) = aH*TN(PiLi/yjPjLj) (5) where NiH is province i's fiscal need for health, xH-0.090 is the weight assigned to health, TNH is the 30 provinces' total fiscal need for health, Li is the ratio of the national average life expectancy to that in province i, PILi/y-PjLj is the share of province i's fiscal need in the 30 provinces' total need for health. For social welfare (k=S), the fiscal need of province i is calculated using the following formula: 147 ANNEX 11 NiS = TNS(O.5*PiOLDi/EjPjOLDj + 0.5*PiUNMi/YjPjUMPj) = aS*TN(0.5*PiOLDi/ljPjOLDj + 0.5*PiUMPi/:PjUMPj) (6) where NiS is province i's fiscal need for social welfare, aS=0.031 is the weight assigned to social welfare, TNS is the 30 provinces' total fiscal need for social welfare, OLDi is the ratio of the percentage of elderly population (over age 65) in province i to the national average, UMPi is the ratio of the urban unemployment rate in province i to the national average, and 0.5*PiOLDi/YjPjOLDj + 0.5*PiUMPi/YjPjUMPj is the share of province i's fiscal need in the 30 provinces' total need for social welfare. For government administration (k=G), the fiscal need of province i is calculated using the following formula: NiG = TNG(Pi/yjPj) = uH*TN(Pi/ZjPj) (7) where NiG is province i's fiscal need for government administration, aG=0.219 is the weight assigned to government administration, TNG is the 30 provinces' total fiscal need for government administration, and Pi/YjPj is the share of province i's fiscal need in the 30 provinces' total need for government administration. For police and law enforcement (k=P), the fiscal need of province i is calculated using the following formula: Nip = TNp(PiLi/FjPjLj) = ap*TN(PiUBi/YjPjUBj) (8) where Nip is province i's fiscal need for police and law enforcement, uxp=0.072 is the weight assigned to police and law enforcement, TNp is the 30 provinces' total fiscal need for police and law enforcement,UBi is the ratio of percentage of urban population in this province to the national average, andPiUBi/YjPjUBj is the share of province i's fiscal need in the 30 provinces' total need for police and law enforcement. For infrastructure (k=I), the fiscal need of province i is calculated using the following formula: Ni = TNI(0.5*LRilELRj + 0.5*Ai/EjAj) = cl*TN(0.5*LRi/LRj + 0.5*Ai!YjAj) (9) where Nil is province i's fiscal need for infrastructure, aG=0.077 is the weight assigned to infrastructure, and TNj is the 30 provinces' total fiscal need for infrastructure. LRi is the total length of roads in province i, and A, is the area of province i. The former reflects the need for maintenance, and the latter reflects the cost due to sparsity of population. 0.5*LRi/YLRj + 0.5*Ai/ZjAj is the share of province i's fiscal need in the 30 provinces' total need for infrastructure. For other services (k=O), the fiscal need of province i is calculated using the following formula: Nio = TNo(Pi/YjPj)= aO*TN(Pi/YjPj) (10) where Nio is province i's fiscal need for other services, aO=0.234 is the weight assigned to other services, TNO is the 30 provinces' total fiscal need for other services, and PI/YjPj is the share of province i's fiscal need in the 30 provinces' total need for other services. 148 ANNEX 11 Step 3: summing up province i's needs in the seven categories to get the total fiscal need of the province: Ni = aE(PiEi/YjPjEj) + aH(PiLi/jPjLj) + xS(O.5*PiOLDi/ZjPjOLDj + 0.5*PiUMPi/EjPjUMPj) + (XG(Pi/YjPj) + up(PiUBi/IPjUBj) + q(0.5*LRi/Y-LRj + 0.5*Ai/EjAj) + ctO(Pi/ZjPj) (11) where N1 is the total fiscal need of province i. Equation (11) can be rewritten as follows by combining some terms: Ni aE(PiEi/-jPjEj) + xH(PiLi/1jPjLj) + aS(0.5*PiOLDi/1jPjOLDj + 0.5*P1UMPi/-jPjUMPj) + ap(PiUBi/ZPjUBj) + X(0.5*LRi/YLRj + 0.5*Ai/1jAj) + (aG+*O)(pi/Zjpj) = 0.276(PiEi/1jPjEj) + 0.090(PiLi/jP)jlj) + 0.31(0.5*PiOLDi/YjPjOLDj + 0.5*PiUMPi/jPjUMPj) + 0.072(PiUBi/YPjUBj) + 0.077(0.5*LRi/YLRj + 0.5*Ai/EjAj) + 0.453(Pi/1jPj) (12) Step 4: adjusting for cost differentials across provinces. The above calculation has not considered the cost differentials across provinces in providing the same level of public services. With limited data, a wage-and-cost index was constructed, using prices of food and construction materials and wage levels. Each of the first two commodities is given a weight of 0.25, and the wage level is given a weight of 0.5. There is obviously much room for improvement, but the present data are sufficient to serve as an illustrative example. The wage-and-cost index is fixed at I for the national average. If a province's index is higher than 1, it means the unit cost of providing public services there is higher than the national average, and vice versa. The index figures are shown in Table A-1(b). The cost adjusted fiscal need of province i is: ANi = WCliNi (13) where ANi is the cost adjusted fiscal need of province i, WCli is the wage-and-cost index of province i, and Ni is the fiscal need calculated using equation (12). Transfers to the Provinces Using two different definitions of "needs", the transfers to the provinces are different. If the fiscal need figures (unadjusted for cost differentials) are used, the transfer from the center to province iis: Ti = Ni - Ci 149 ANNEX ]1 where Ni is given by equation (12). If the cost adjusted fiscal need figures are used, the transfer from the center to province i Is: T]= ANi -C where ANi is given by equation (13). Transfers calculated using these formulas are presented in Table Al 1.3. For comparison, the actual transfers in 1994 are also presented in Table Al 1.3. The above calculations assume that 100 percent of the central government transfers made in 1994 be allocated according to the proposed equalization formula. In other words, the proposed system is a "full" equalization system. However, the distribution of transfers under this system is distinctly different from the actual allocation in 1994: the standard deviation (the average difference between the proposed per capita transfer and the actual per capita transfer) is 263 Yuan, or 198 percent of the average per capita transfer in 1994 (See Table A-4). Obviously, such a drastic reallocation of resources is politically infeasible. Two alternative "partial" equalization schemes were, therefore tried: (1) 50 percent equalization. 50 percent equalization means that 50 percent of the actual central government transfers (net) made to the localities in 1994 are allocated in proportion to the original allocation and the other 50 percent are allocated by the proposed equalization formula using cost adjusted figures. From Table Al 1.4, one can see that under this system the standard deviation from the actual allocation now becomes 132 Yuan or 98 percent of the average per capital transfer in 1994. (2) 20 percent equalization. 20 percent equalization means that 80 percent of the actual central government transfers (net) made to the localities in 1994 are allocated in proportion to the original allocation and the other 20 percent are allocated by the proposed equalization formula using cost adjusted figures. From Table Al 1.4, one can see that under this system the standard deviation from the actual allocation now becomes 53 Yuan or 39 percent of the average per capital transfer in 1994. Does the Transfer System Equalize? The transfer system designed above aims to equalize the provinces' abilities to provide public services at similar levels of tax effort. While equalizing per capita income is not the direct objective, due to a high correlation between income and fiscal capacity, a transfer system like the one suggested above should also have strong redistributive effects on per capita income. 150 ANNEX 11 Table A11.3: Fiscal Needs and Fiscal Transfers Unadjusted for cost differentials Adjusted for cost differentials - Fiscal Formula Per capita Fiscal Formula Per capita Actual need transfer formula need transfer formula per cap transfer transfer transfer (MiL. Y) (MiL. Y) (y) (Mil. Y) (Mil. Y) (y) (y) All China 392962 161803 135.0 387677 161803 135.0 135.0 Beijing 3735 -5238 -465.6 447 -4687 -416.6 468.3 Tianjin 3014 -1525 -163.1 3086 1502 160.6 237.1 Hebei 19585 10030 157.0 19315 10089 157.9 102.7 Shanxi 9640 4998 164.2 7833 3327 109.3 116.3 inner Mongolia 9513 6226 275.5 7809 4700 208.0 250.1 Liaoning 13300 633 15.6 12381 -281 -6.9 171.9 Jilin 8634 3368 130.8 7528 2354 91.5 207.1 Heilongjiang 12982 3557 -96.9 10688 1343 36.6 157.2 Shanghai 4513 -9657 712.2 5825 8644 -637.5 156.5 Jiangsu 21493 3529 50.3 22367 4535 64.6 90.5 Zhejiang 13285 -135 -3.1 16199 2825 65.8 136.0 Anhui 19400 12933 217.2 16936 10857 182.3 64.8 Fujian 10550 3625 113.9 12613 5843 183.6 143.9 Jiangxi 13114 8799 219.2 11341 7288 181.5 106.5 Shandong 26869 9977 115.1 24199 7593 87.6 97.0 Henan 27687 17421 193.0 23007 13240 146.7 84.5 Hubei 18444 8628 150.9 17340 7792 136.2 104.5 Hunan 19790 11505 181.0 20149 12253 192.8 103.2 Guangdong 20700 -3860 -57.7 32586 8101 121.1 176.6 Guangxi 14243 8683 193.2 15363 10110 225.0 139.5 Hainan 2459 1427 200.8 3021 2047 287.9 175.5 Sichuan 36252 23695 211.3 32657 20826 185.7 90.4 Guizhou 11963 9487 274.3 11436 9267 268.0 124.3 Yunnan 14559 6820 173.1 13794 6269 159.1 322.5 Tibet 3374 3129 1325.7 3961 3830 1622.8 1049.2 Shaanxi 11045 7260 208.6 10129 6569 188.7 123.3 Gansu 8987 6350 267.1 8737 6307 265.2 182.1 Qinghai 3139 2516 530.8 2850 2306 486.5 387.1 Ningxia 1810 1188 235.7 1683 1099 218.0 242.3 Xinjiang 8321 5876 360.1 8396 6148 376.7 259.8 Source: Tables A 11.1 (a) and (b), and author's own calculations. 151 ANNEX 11 Table A11.4: Fiscal Transfers under Alternative Equalization Schemes (Yuan) 100% Equalization 50% Equalization 20% Equalization Actual Per cap Differ. Per cap Differ. Per cap Differ. per cap transfer from transfer from transfer from transfer act. amt. act. amt. act. amt. 1994 All China 135.0 0.0 135.0 0.0 135.0 0.0 135.0 Beijing 416.6 -884.9 25.8 -442.4 291.3 -177.0 468.3 Tianjin -160.6 - 397.7 38.2 -198.9 157.6 -79.5 237.1 Hebei 157.9 55.2 130.3 27.6 113.8 11.0 102.7 Shanxi 109.3 - 7.0 112.8 -3.5 114.9 -1.4 116.3 Inner Mongolia 208.0 -42.1 229.0 -21.1 241.7 -8.4 250.1 Liaoning -6.9 -178.8 82.5 -89.4 136.1 -35.8 171.9 Jilin 91.5 -115.7 149.3 -57.8 184.0 -23.1 207.1 Heilongjiang 36.6 -120.7 96.9 -60.3 133.1 -24.1 157.2 Shanghai -637.5 -793.9 -240.5 -397.0 -2.3 -158.8 156.5 Jiangsu 64.6 -25.9 77.6 -13.0 85.3 -5.2 90.5 Zhejiang 65.8 -70.2 100.9 -35.1 122.0 -14.0 136.0 Anhui 182.3 117.5 123.6 58.8 88.3 23.5 64.8 Fujian 183.6 39.7 163.7 19.8 151.8 7.9 143.9 Jiangxi 181.5 75.1 144.0 37.5 121.5 15.0 106.5 Shandong 87.6 -9.4 92.3 -4.7 95.1 -1.9 97.0 Henan 146.7 62.2 115.6 31.1 96.9 12.4 84.5 Hubei 136.2 31.8 120.4 15.9 110.8 6.4 104.5 Hunan 192.8 89.6 148.0 44.8 121.1 17.9 103.2 Guangdong 121.1 -55.5 148.9 -27.7 165.5 -11.1 176.6 Guangxi 225.0 85.5 182.3 42.8 156.6 17.1 139.5 Hainan 287.9 112.4 231.7 56.2 198.0 22.5 175.5 Sichuan 185.7 95.3 138.1 47.6 109.5 19.1 90.4 Guizhou 268.0 143.7 196.2 71.8 153.1 28.7 124.3 Yunnan 159.1 -163.4 240.8 -81.7 289.8 -32.7 322.5 Tibet 1622.8 573.7 1336.0 286.8 1163.9 114.7 1049.2 Shaanxi 188.7 65.4 156.0 32.7 136.4 13.1 123.3 Gansu 265.2 83.1 223.6 41.6 198.7 16.6 182.1 Qinghai 486.5 99.4 436.8 49.7 407.0 19.9 387.1 Ningxia 218.0 -24.2 230.1 -12.1 237.4 -4.8 242.3 Xinjiang 376.7 116.9 318.3 58.4 283.2 23.4 259.8 SD 263.5 131.8 52.7 C.O.V. 1.95 0.98 0.39 Note: 100 percent equalization means that 100 percent of the actual central government transfers (net) made to the localities in 1994 are allocated by the proposed equalization formula using cost adjusted fiscal need measurements.50 percent equalization means that 50 percent of the actual central government transfers (net) made to the localities in 1994 are allocated in proportion to the original allocation and the other 50 percent are allocated by the proposed equalization formula using cost adjusted figures. 20 percent equalization means that 80 percent of the actual central government transfers (net) made to the localities in 1994 are allocated in proportion to the original allocation and the other 20 percent are allocated by the proposed equalization formula using cost adjusted figures. SD: standard deviation; C.O.V.: coefficient of variations. 152 ANNEX 11 The following regression is conducted to test the hypothesis that a transfer system equalizes per capita income across provinces : PCTi ao+ aI PCGDPi where PCT, is the per capita transfer to province i, and PCGDPi is the per capita GDP of province i. If alis negative and statistically significant, it means that the system has a significant equalization effect. When per capita transfers are calculated using need figures unadjusted for cost differentials, al is significantly negative. At the same time, the R-square is 0.55, implying that 55 percent of the variations of the transfers across provinces serves the purpose of "equalization." This indicates a significant redistributive effect of the proposed transfer system. The regression results are as follows: Regression III: PCTi= 544.26 - 91.26 PCGDPi (2.55) (-14.51) R-square = 0.59, No. of observations = 30, Degrees of freedom =28. When per capita transfers are calculated using needs figures adjusted for cost differentials, al is also significantly negative. The R-square is 0.42, showing a slightly weaker correlation between the two indices than that in Regression III. The regression results are as follows: Regression IV: PCTi = 518.77 - 83.93 PCGDPi (1.93) (-4.54) R-square = 0.42, No. of observations = 30, Degrees of freedom =28. For comparison purposes, the actual transfer figures in 1994 were used to run the same regression. The resulting al is statistically insignificant and the R-square is only 0.0002, showing not even a slight correlation between per capita transfers and per capita GDP levels. This suggests that the current transfer system has not effectively achieved any redistributive goal. The regression results are as follows: Regression V: PCTi= 206.46 - 0.97 PCGDPi (1.10) (-0.075) R-square = 0.0002, No. of observations = 30, Degrees of freedom =28. The same regression is also conducted using the per capita transfer figures generated by the "partial equalization" transfer schemes proposed in the previous section. Table Al 1.5 compares the regression results of four systems: actual transfers in 1994, 20 percent equalization, 50 percent equalization, and full equalization. The results show that the full equalization and 50 percent equalization schemes have statistically significant equalization effect (the slope coefficients are significantly negative). The 20 percent equalization schemes do have some equalization effect (the slope coefficient is negative), but it is not statistically significant. The actual transfer scheme has the least equalization effect (the slope coefficient is almost zero). 153 ANNEX ]] Table A11.5: Regression Results under Four Transfer Schemes: PCTi ao + al PCGDPi 1994 Actual 20% 50% 100% Transfer Equalization Equalization Equalization ao 206.46 268.92 362.62 518.77 (1.10) (1.36) (1.66)* (1.93)* a -0.97 -17.4 -41.95 -83.93 (0.075) (-1.30) (-2.81)* (-4.54)* R2 0.00 0.06 0.22 0.42 DF 28 28 28 28 Note: The calculations use cost adjusted figures on fiscal needs. Numbers in parentheses are t ratios. A "*" indicates that the t ratio is statistically significant. Conclusions Statistical evidence suggests that China's current fiscal transfer system perform almost no redistributive function. The illustrative example of a formula-based equalization transfer model presented in this Annex shows that an important improvement in redistribution can be made by introducing appropriate measures of fiscal capacities and needs with appropriate variables. One should notice that shifting from the current transfer system to a "full equalization" system might not be feasible in the short- or medium run. A pragmatic approach is to increase the magnitude of the new transfer scheme gradually over time in order to minimize political difficulties. The purpose of this illustrative example is not to provide the exact model that China is to use; the intent is to offer an alternative methodology to the ones that are being considered. 154 ANNEX 12 ANNEX 12: FINANCIAL MANAGEMENT INFORMATION SYSTEMS Information is the lifeblood of budgetary, resource allocation and financial management. Financial management information systems (FMIS) provide decision-makers and public sector managers with a set of tools to support: 1. Controlling aggregate spending and the deficit; 2. Strategic prioritization of expenditures across policies, programs and projects for allocative efficiency and equity; 3. Better use of budgeted resources, i.e.; to achieve outcomes and produce outputs at the lowest possible cost. China's budgetary reforms will need to pay attention to further developing information systems. Much of the recommendations made in the report cannot be implemented without sufficient information to support budgetary decisions and implementation monitoring. China is still at the start of developing the kind of information system that is discussed in this annex. Developing of such a system will take considerable time and resources, but should be taken serious, if budget management is to be improved. Improving systems and processes requires an understanding of how these three levels interact. Particular attention needs to be given to the capacity to link the three levels through a set of commonly shared databases. There is a need for vertical integration-through the planning, budgeting and financial management cycle-and horizontal integration-accounting, budgeting, cash management and audit. This annex discusses approaches to this integration. Traditionally, control has been the dominant, if not the single, reason for developing FMIS'. This control role will always be central. However, to deliver on the three levels of performance, a FMIS that supports a result orientation at the strategic and operational level will be needed. Financial data that supports informed decisions on policies and programs and that link information on costs, outputs and outcomes will be essential. The trend to decentralize within and between levels of government also underscores the importance of adequate FMIS. This annex is not advocating a big bang theory of FMIS. A modular approach will often be appropriate. Most of the elements of a comprehensive system exist in some form in every country. The nature and pace of integration of these elements will vary from country to country, but each step must take into account the end point-timely and reliable information for use throughout the resource allocation, budgeting and financial management cycle. Some countries may be in a position to take advantage of developments in information technology to move relatively quickly from a currently disabled, manual system to an automated, integrated system. Automation will not, however, resolve problems associated with structure and a lack of discipline on the part of decision-makers. 155 ANNEX 12 FMIS INADEQUACIES Inadequacies in FMIS show up in all aspects of budgeting, resource allocation and financial management. From the perspective of the three levels, the most obvious symptoms of inadequate systems are: 1. At the macro level, assessment of the fiscal situation is derived from the books of the central bank; 2. At the strategic level, there is no costing information on policies and programs; 3. At the operational level, financial information is not available on time or in a form that facilitates effective expenditure control or management of agencies, programs or projects. More generally, the symptoms include a lack of timely and appropriate financial information for decision making during planning and budget formulation; lack of timely and accurate financial management information during budget execution; and non-existent, or out-of-date, financial reporting. Some of the factors underlying these inadequacies include: * Gaps in the legal framework * Lack of a standard classification coding structure * Poorly specified reporting requirements-external and internal * Multiple ex ante controls on expenditure * Multiple FMIS, often computer based, across government that do not communicate with each other * An inability to move data from one stage in the budgetary cycle to the next. INSTITUTION BUILDING AND POLICY REFORMS Implementing and improving information systems should be seen in the context of wider institution building and policy reform efforts. FMIS influence, and are influenced by, an overall regulatory framework that consists of: a. The control structure that governs the use of government funds and are derived from the legislative framework; b. Accounts classification, which enables consistent recording of each financial transaction for expenditure control, costing, and economic and statistical analysis; and, c. Reporting requirements covering: (a) external reporting-to provide information to the legislature and the public, as well as to other countries, international organizations, overseas investors and financial markets; and, (b) internal management reporting for government policy makers and managers. Computerization is a key issue linked to any consideration of FMIS. As a minimum, the adequacy of the regulatory framework will need to be reviewed, and modified as appropriate, and the institutional capacities of the responsible government agencies analyzed-inadequacies highlighted and corrective measures identified-before computer systems can be designed to support fiscal management. The benefits realized from computer systems will depend on the degree of success achieved in strengthening basic 156 ANNEX 12 functional processes, the regulatory framework and the organizational structures responsible for them. Computerization is discussed later in this Annex. DEVELOPING A CORE SYSTEM The central premise of this annex is that the broad technical requirements of government budgeting and accounting are similar for all countries, and a core system can be specified to meet these requirements. Such a system would meet basic needs for most countries, but aspects would need to be tailored to suit particular needs. The rate of implementation would depend on the capacity of individual countries. Better integration of the subsystems is the ultimate objective, so it is important to build in this capacity from the outset. Many countries could start with a core accounting system that consists of modules for accounts payable, accounts receivable and the general ledger. An automated accounting system would ensure completeness of data capture. No transaction would be processed outside the system, and rigorous financial controls would be applied to all transactions processed by the system. With the core accounting system as a foundation, government could then expand the system as capacity developed. Developing a system involves: (a) analyzing business processes involved in fiscal management; (b) defining a general information architecture, which is derived by an analysis of the strength of information linkages among the business processes; and (c) defining a general system architecture, based on the information architecture. The general system architecture can be used for the design and development of application software, or for the selection of existing software packages that meet requirements. The functional processes carried out by the central government in the areas of budgeting and accounting-and linkages to the control framework-are illustrated in Figure A12.1. As indicated, the functional processes can be categorized as those carried out by the central agencies and those carried out by the spending ministries and agencies. Those of the former group are most directly linked to the control framework. Indeed, one of the main functions of the central agencies is to ensure that the control framework is properly applied throughout government. Figure A12.2 shows the core elements of the information system network required to surport government fiscal management and the main information flows between these elements. Analyzing data from a number of countries on fiscal management (FM) processes has derived this information architecture and the information systems supporting these processes. The Y-axis lists the main processes in FM, the X- axis the organizations normally responsible for these processes. Each box lies at the intersection of the functional process and the organization(s) normally responsible for the process and is the information support system for that process. Each system could comprise a number of subsystems. Figure A12.2 also shows the main information flows between the system modules. FM support at macro, strategic and operational levels Macroeconomic forecasting. The MOF, the Planning Organization, and the Central Bank normally use this group of systems to support the setting of fiscal policy. The systems assist with macro fiscal planning and the development of the macroeconomic framework. This is in turn used by the MOF to set aggregate budget parameters and guidelines for budget agencies to submit estimates. These systems require data from external economic databases and the assumptions regarding GNP, inflation rates and the deficit. 157 ANNEX 12 t I unciInalr1 I :.r.r,.l rentrail P'ri.nesw.11 ntr'l n ILJCIr L r ind 4 . h 1In t.itrr.Lnr PuUil ;est.r 1 r-rk I'r-cr.am 11.Anigc*menu- Macroeconsom.ic Budget Circular Framework BudgetPrpsl Fund Structure Recurrent Capital Existing Programs Organic Budget & Projects Law New Proposals Appropriation Approved LAW Budget Revenue Projections Troposed Work Program Supplhmnt-try Consolidated Cash Cash Requirements Appropriat,i Law - Flow Forecasts Forecasts arrant Releases Fund Requests to Ministries Financial Regulations Accounting Purchase Orders System Purchase Contracts/ Commitments ccoutsGoods Receipt payable and Verifications Check Voucher Payment Vouchers Issue Payment Orders to Bank Reprtin Treasury General Agenc.y . ..neral Reuircinnt Ledger System Ledger System Tax/Nontaxl Receipt Transfers to Receipts& Loan Treasury Account Receivabke Tax and Nontax Receipts .Issues and Redemptions of Govt. Securities .Reconcilaion Revisions to Budgv(Rev iews Revenue Projections & iscal Reports I&Work Programs 158 ANNEX 12 Figure A12.2: Information Systems Architecture,for Government Fiscal Managemnt Function Central MOF Planning Ministry of Finance Regional Government Spending Public Revenue Coll. AgenciesPaying/Recv Audit Bank Debt Mgmt Organ. Bud. Mgmt P.E. Monitoringash Mgm ccounting Treasury Units Agencies Enterprises Taxes Customs Banks Org. Macro Legend 0 Arrows show information flows. Full lines indicate Economic System for MaCro-Economic Forecasting electronic flows, dotted lines, paper based flows Forecasting Key Information Systems Modules for GFM Macro-Economic Framework Dataon SpendingAgenc usr. Enver T,ax Uto previous Btid Prep BinP.p Systemn System andcurrent Systems s Feedback year Budget Guidelines Budget Guidelines from actuals Audit Budget Systerns for Budget Budget Proposals to various Preparation A Historcal Preparation areas Baseline data Investment Curren Revenue Estimates 4 1udgeted Figpres on Projects Budget Investment Carrent Budget Appropriations : Spending ub Enmrpns 4 , ... . ,Agency BudEsec Budget Systems for Vr,r3 t Core Government ud Exec Syst [ im J Execution, Budget Execution Accounting ---aymentsn, Accounting and System " " Rece4pt and Fiscal St. of Borrowings Fiscal Reporting Bank. ..,. r..Reconcilation Reporting nd Public Debt Interest pror m san pro ts Paying tlen Is , , *nd,ng Age-nCr Recelvinj Monitoring * .r...>n ri.oirnentPro.ers Bank tvestment &sv utnn,un Systems Pr ,itar-.1 1 . . . I . :. u . . 1 r e . Cash Cash Requirements Cash ash 4Epen fi es Returns on key indicators from Public Enterprises Management Mgmt i Sofowing Re4uirement System a sh Allocations Debt Debt Managomen For .qn Foreign Assistance Utilization (on investment proj3cts) Management Systems As,%.%Idnr Foreign Assistance Disbursements Ori)nel.c I F-reqn 06rdInl'12i Debt Service .ayments/Loan receipts [ a x Custors Revenue Admins Admin Admin. y Stem svuems o Payroll & Pension u.. Personnel Civil Service Post - Systems Information Management Authorized posts its.ageme Systems Audit . !...v-v tfor Input to Audit Systems from Various Areas GENSYSAUI 159 ANNEX 12 Additionally, they require information on programs and projects the government intends to implement over the period of the program such as data on tax and non-tax revenues, and data on domestic and external borrowings, which are maintained by other components of the FMIS network. Budget Preparation and Approval. These information systems support strategic priority setting by assisting with the preparation of budget estimates. The systems receive details of ongoing and planned programs and projects from line agencies, consolidate them and produce from them the documents that form the basis of the negotiations between the line agencies and the MOF. After finalization of the budget, the systems produce the approved budget estimates. The systems should be able to capture and maintain the budgetary proposals and income estimates of all government agencies and to capture subsequent changes during the budget preparation, approval and amendment processes. The evaluation of the budget proposals include an examination of the manpower component, maintenance and other operating expenses, and the evaluation of the capital outlays program, using baseline data from previous periods for comparison. Examination of the capital budget requires data on the physical and financial status of government-approved projects, both locally and foreign funded. The system should be able to access and generate the baseline data from the relevant past-year databases. Budget execution, core accounting and fiscal reporting. These systems support operations and are the primary repository of financial data that serve as the basis of the government's FMIS. These systems perform basic accounting functions, processes for budget execution, monitoring and control, and provide the information required for cash management and to implement cash limits in order to obtain the status of actual expenditures on ongoing projects. These systems also monitor and evaluate the overall budget implementation processes and produce the necessary fiscal reports. In addition, these systems would provide useful financial information to the line ministries and spending units to enable them to better manage their work programs. These systems need to be comprehensive in coverage and be a source of reliable and timely data to become a credible source of information for users. These systems maintain data on: a. approved (capital and recurrent) budgeted appropriations, b. sources of financing for programs and projects, c. budget transfers, d. supplementary allocations, e. fund releases (warrants) against budgetary allocations over the course of the year, and f. data on commitments and actual expenditures against budgeted allocations. The budget execution systems normally operate at two levels-at the line agencies and at the MOF. The line agencies' systems, operated by their finance departments, enable managers to track the budget implementation process and implement expenditure controls at the agency level. The central systems track the budget execution process for the government as a whole. At the start of the fiscal year, the legislature-approved budget is entered into the system. Budget transfers such as supplementary authorizations and warrants are also entered during the year they are issued. As commitment and expenditure transactions take place at the line agencies, these systems: (a) check for budget authorization and the existence of a prior commitment, (b) record information verifying receipt of 160 ANNEX 12 goods, (c) authorize payment, and (d) update the total amounts committed and spent. They operate on the basis of commitment, verification and payment request transactions received from the line agencies, either electronically or on paper. Subsystems. There are a number of subsidiary systems that are essential to a well-performing public sector. Subsidiary Payment Systems. These systems cater to payments such as payroll and pension. The systems post summaries to and interface with the core accounting system. Cash Management System. This system maintains an up-to-date picture of the government's liquidity position and cash requirements. This system obtains information on actual agency expenditures and cash balances in government (including agency) accounts from the general ledger, revenue inflows, borrowing, loan disbursements, treasury bills, government bonds and cash deposit maturities. This information is obtained either from the general ledger or from the systems for these areas, e.g., the debt management system. Government can use this information to decide on: (a) budget ceilings and fund releases to line agencies; and (b) the timing of the issues and redemption of government securities to provide short-term financing for shortfalls. Debt Management System. This system maintains information on public domestic and external borrowings. This includes information contained in loan documents and transactions and issues of government securities. In addition to accounting information, these systems also provide important information required in the formulation of fiscal policy such as forecasts of draw-down and debt- servicing liabilities and debt implications of fiscal and deficit financing policies. Civil service management systems. This system assists in aspects of civil service management that are relevant to financial management (FM). These are processes associated with post management and payroll and pension payments. The corresponding system modules therefore form important elements in the FMIS network. Revenue Administration. This group of systems assist the government in: (a) the processes for the formulation of tax and tariff policies, and (b) the collection of tax and non-tax revenue. The tax revenue administration system provides summary information on revenue collections to the core accounting systems. Auditing. These systems assist the internal and external audit functions. Auditing takes place at two levels: internal audit at the line ministries during the course of the FY and external audit by the auditor general through random checks and on the final accounts for the FY. INTEGRATING INFORMATION SYSTEMS Integration of the different elements of the FMIS network will not happen without conscious effort and the involvement of all stakeholders. More often than not, information systems are implemented as components of separate projects, responding to specific needs, with little thought given to requirements in other areas and or to critical interrelationships. The resulting information systems are often disparate and segmented with little or no capacity for sharing data. These systems have overlapping and sometimes conflicting functions and provide incomplete coverage, particularly for managerial information requirements, which normally span several areas. 161 ANNEX 12 The failure to integrate financial management information results in: * fragmented and unreliable data, * duplication of data difficult to reconcile, * failure to use actual results in planning and budgeting, * failure to fully and publicly report financial and operational results, and * undue emphasis upon one of the component subsystems, usually budgeting, which tends to dominate, duplicate and crowd out the others. The investments required to set up modern computer-based information systems coveringthe major areas of FMIS are sizable and can span several years. Investments could easily range from US$10 to 50 million over five years. Integration is a requirement, not an option, for these large investments to yield expected benefits. Incompatible or duplicative systems would be wasteful and inefficient and would not provide the information that governments require for economic management. Components of an IFMS An integrated financial management system (IFMS) consists of an interrelated set of stbsystems that plan, process and report upon resources, quantifying them in financial terms. The basic subsystems normally are accounting, budgeting, cash management, debt management and related internal controls. An important element of modern internal control throughout government is a professional internal audit function that is an integral part of an IFMS. The factor that "integrates" the system is a set of commonly shared, reliable databasess to and from which all data are expressed in financial terms flows. All of the subsystems and users of financial data must participate in common data sharing. The validation, classification and recording of data in the accounting subsystem produces timely reports of classified data for users. The common database may be manually maintained, but an electronic database is much more flexible and accessible. An IFMS can be developed regardless of the organizational structure, but it is likely to function better where the four basic subsystems are closely related under a common, professionally qualified financial management executive. In integrated networks, the component elements are able to exchange information with ease and to share commonly used data. Integrated systems must be structured along functional rather than organizational I ines. The system needs to support a functional area across all organizations being integrated. This enables the creation of systems and databases in which the agency responsible for a function provides a subset of data. However, the databases are accessible by relevant core agencies, subject to appropriate security controls. All agencies work with the same set of data, eliminating duplicative data gathering and data inconsistencies. 162 ANNEX 12 Framework for integration The first step towards achieving integration is to develop a framework that provides an overview of the system network required to support FM. The framework is developed by analyzing: (a) basic functional processes and information requirements; (b) responsibilities of agencies in charge of the processes; (c) information flows between the processes, including the nature, volume and frequency of these flows; and (d) data characteristics of the information used and created by the processes. This framework would address questions such as: a. Which information system modules are required to support FM; b. What is the scope, scale and type of a particular system component; and, c. How do these systems interrelate? The framework would consist of: a. A system architecture that identifies the major component modules of the network required to support FM, the type of information maintained by each module and the information flows between modules; and b. A technology architecture that identifies the appropriate technology clDices for the hardware and software to set up the modules. The integration of system modules can be achieved incrementally once the framework has been set up and the prerequisites and criteria for integration have been spelled out and incorporated in the implementation plan. The framework will serve as a road map for implementation. Automation Fiscal management processes are transaction intensive. Retrieving information from manual records and reclassifying it in an appropriate format or classification can be extremely time consuming and labor intensive. Automated systems speed up the process and provide the accurate data required for economic and fiscal management. Three aspects of computer-based information systems are particularly important: a. The rapid compilation of data on transactions from widespread locations and timely generation of information in formats prescribed by the legislature or for internal control; b. Ensuring that prescribed controls and procedures are adhered to at the point of origin of a transaction; and c. The integration of the posting and classification of transactions with the functional process in such a way that performance (e.g., authorizing a payment) simultaneously ensures that all transactions are classified and posted. This in turn will ensure complete and timely data. 163 ANNEX 12 Technological recommendations. Three important technological improvements for a well- performing information system are: (a) multi-tiered networks, (b) system portability and scalability, and (c) application-specific technology. Multi-tiered networks. Several elements of the FMIS network require system modules at the line agency and central levels with facilities for generating, storing, and processing data at each level and for exchanging data between levels. The data volumes encountered can vary widely across the nodes of the network. Govemment-wide FMIS requires a multi-tiered network. This could consist of stand-alone microcomputers, local area network (LANs) or minicomputers, located at the nodes (MOF, other core agencies, the line agencies and subordinate/regional treasury offices) and connected by telecommunication lines. Local computers carry out the transaction processing and database management at each node. The summary or detailed data required for the applications are transmitted to the computer in the agency responsible for that system (e.g., to the MOF's budget division for the budget system, to the treasury for the accounting and cash management systems). This configuration is often preferred for two reasons: (a) computing power is distributed commensurate with node requirements, making this system less vulnerable to malfunctions at the central site; and (b) end users at the line agencies have more control of technological and data resources, which inculcates a sense of ownership in the system. In the absence of good telecommunication facilities, the data transfer between the nodes and the center could be periodic (daily, weekly or monthly, depending on the application system) in an off-line/batch node. The size of each node's computers would depend on the amount of data and number of transactions. They could be stand-alone microcomputers, microcomputers connected by a LAN, or fairly large capacity minicomputers at the center and larger line agencies. Systems portability and scalability. For system modules implemented at multiple levels, the software at each node should be able to be run on small or large computers without major changes. These properties can be achieved by choosing compatible computers that offer multiple size configurations. However, this would restrict additions to the network to this vendor and line of computers. To avoid these restrictions, the application systems should be developed using tools and DBMS portable software that can operate on machines of different sizes offered by several vendors. To ensure vertical and horizontal portability, and scalability, the hardware should be an open system-assembled from components that conform to generally (though not universally) accepted standards. The hardware and software would therefore be interchangeable, providing greater flexibility. It will be some time before there is a full set of products on the market that truly conforms to open system standards; at present, the UNIX environment comes the closest. Most vendors now offer a version of UNIX.. Since UNIX versions vary slightly with the vendor, some application changes may be required before it can be used on a different vendors' machine; however, time and effort involved in making these changes would be small compared to rewriting the applications. Certain tools such as fourth-generation languages (4GLs), RDBMs and graphic user interfaces make it easy to add or change application features, including changes to data-base structures, associated programs and report formats. The use of these tools increases application 164 ANNEX 12 development productivity and, therefore, reduces development time. These tools also enable end users to access the databases and to program simple reports. Application-specific technological requirements. The technology required for the system modules would depend on the modules' functional characteristics, including the amount of data handled, the size of the data-bases, the number and rates of transactions, and the volume of the information flow between modules. The distribution of information processing, either centrally or among widely separate locations, is important. If distributed among widely separate locations, the following factors should also be taken into account: how frequently the information needs to be aggregated at the center, and the requirements for output facilities such as graphics, report writing and desktop publishing, and for analytical facilities. Off-the-shelf commercial packages are capable of filling many of the financial management needs of government-general ledger, accounts payable and receivable. For all currently known configurations, however, such packages will require substantial additional work to meet the complex central budgeting and accounting needs. There is a great body of evidence that each country will have its own idiosyncrasies that will produce a design that is unique to that country. E. MANAGING IMPLEMENTATION Implementation of budget and accounting system reforms is a long-term process. Such programs have been evolving in the OECD economies for several decades. A similar or longer-term timetable will be necessary in countries with less developed administrative processes. For effective resource allocation, budget and financial management, systems must be developed around the core functions described in the preceding sections. Moreover, since the underlying requirements are similar, it may be feasible to draw on the growing number of off-the-shelf software packages to develop a core government budgeting and accounting system (CGBAS) that could (a) provide the basic transaction processing and database management functions needed for fiscal management in any country; and (b) allow the tailoring of controls, data entry and reporting functions to any country's needs. However, off-the-shelf software packages should not be regarded as the "off-the-shelf solution." As a minimum, they will require significant modification to accommodate the unique demands of a government system of resource allocation, budgeting and financial management. In fact, experience has shown that because of the unique set of institutional arrangements in each country, the design of the FMIS is unique to that country. The case for developing a core system or identifying suitable software packages is strongest for the EITs. Development of the appropriate institutions to manage a market-oriented economy is of strategic importance to these economies. They are all starting from a similar administration base, so a core system should be readily applicable to all. Development of a CGBAS should greatly reduce the implementation time for all such systems. The CGBAS should provide a sharper focus for much of the technical assistance provided by international bilateral agencies. The IMF, the World Bank, the regional development banks and the UNDP have provided assistance to improve various aspects of budgeting and accounting. The need for continuing efforts in this area is widely recognized. If a CGBAS were available, its features could be tailored to meet specific needs, and the implementation time would be much reduced from the broader approaches now being used. General support would continue to be needed. In each country, there would need to be a 165 ANNEX 12 comprehensive review of system requirements, administrative constraints and an appropriate phasing of implementation in light of these factors. Centering the efforts on a CGBAS, however, should avoid the difficulties that arise from piecemeal development of systems. Clearly defined requirements would enable the systems to be designed to accommodate all budgeting and accounting objectives, whereas computerization of existing accounting systems would only meet the limited objectives of ensuring financial compliance. Information system development could then play a pivotal role in improving budgeting and accounting, providing that the policy and administrative environment is supportive. Successful implementation of an integrated network of information systems is crucially dependent on cooperation among diverse users. Project preparation and implementation in a multi-agency environment is complex. Key requirements are securing sponsorship at the highest levels of government and participation by the widest range of users in all phases of the project. A necessary condition for sustainability of system reforms is for key decision makers to see information as critical to their work. A steering group with representatives from all major stakeholders should ensure that all of the needs of participating agencies are taken into account during system design. Agencies will then not have to resort to independent and duplicative initiatives. Cooperationamong users would also establish systematic data-sharing arrangements, protocols and schedules between the various systems so that all agencies have access to financial data as needed. The organization in charge of a functional process for the component modules will be responsible for that process. The number of technical staff and skills required to set up such systems are considerable. To ensure sustainability, an information system organization should be established, or existing organizational units strengthened, to incorporate and retain the skills and to manage planning, development and operation. The following skills are required: (a) high-level project design and planning skills; (b) project management skills; (c) technical implementation skills-to operate and use the hardware and software; and (d) user support skills-to develop user and technical documentation and to set up training and a hot line for end users. Information system support would normally be distributed among several agencies aciss government. Therefore, coordinating mechanisms should be created to ensure that a common set of policies, procedures and standards are put in place for managing data and systems across government. The standards should cover: (a) the protocols for communications; (b) data entry; (c) editing and updating screen input and output formats; (d) back-up and recovery; and (e) security, contingency and disaster planning; and technical and user documentation. Project planning methodologies should be used to design, implement and monitor the FMIS, and project management responsibilities should be clearly identified. Phased implementation would ensure that the system could be easily absorbed by organizations. Hardware and software chosen must be supported locally. Vendors represented locally can provide training and technical support needed during the life of the system. 166 ANNEX 12 The links of integration Several "links" are necessary to achieve true integration of the financial management system: training, coordination, harmonization, communication and collaboration. Most financial management systems fail due to the absence of these links. A comprehensive training program covering all aspects of the IFMS is essential to ensure that all individuals involved in the subsystems understand the role, need and underlying concepts involved in each subsystem. All too frequently FM systems fail because budget personnel know nothing about cash management or debt management staff fail to consider accounting requirements. Training staff in each subsystem and other aspects of the system, e.g., internal controls and management principles, would contribute to the success of the system. The coordination and harmonization of the IFMS component subsystems are necessary to achieve "integration." Harmonization is achieved by means of consistent provisions and standards governing application of each of the subsystems. The principles, policies, standards, manuals and procedures must be harmonized to avoid conflicts and inconsistencies between one subsystem and the others. Internal provisions for each subsystem must be coordinated with the other subsystems before being completed and issued in final form. Collaboration among individuals is also important. Communication among staff inthe subsystems must be open. Periodic meetings, conferences or round table discussions should be held to discuss problems, areas of possible friction or duplication and common measures to be taken to achieve more operational efficiency. Information systems normally have a direct effect on the way people do their day-to-day work. Appropriate change management procedures and training need to be carried out to ensure that staff feel comfortable in the new work environment and, in particular, do not feel insecure about their jobs. Information systems may also lead to redefinition of the relative authority and power relationships of individuals and groups within organizations. Change management exercises need to cater to these complex effects also. Partial integration Full integration, like full computerization, may not be cost-beneficial in some countries. Varying degrees of partial integration are viable options. The most important element in any partially integrated approach is the integration of the budgeting and accounting systems so as to assure a commonly used data- base for managerial decision making and for budget formulation. System design should allow for the remaining systems to be integrated and expanded without redesign. Each government can decide on a system after evaluating: the success of existing systems, the need to improve the systems and the opportunity to initiate integration as a part of improvement. The governments can then determine what degree of FMIS integration is reasonable to expect, what expertise will be required and what financing is available. 167 ANNEX 12 Data Architecture for Government Budgeting and Accounting System Business Process Data Entities Definition Created Line agencies prepare budget Line agency budget Description of programs and projects proposed by line agencies for estimates estimates the coming fiscal year and estimates of funds required by budget category Receive budget Approved budget Description of approved programs and projects to be executed by line authorizations from MOF agencies during the year and amount of funds voted after budget is finalized Prepare cash requirement Cash flow forecasts Forecasts of cash requirements for the year based on known and forecast anticipated commitments for both recurrent and capital expenditures Assess liquidity position Liquidity position Status and forecasts of cash requirements and availability made at a specific point in the fiscal year, based on cash flow, forecast revenue projections, foreign aid inflows, and data on maturities of term deposits Issue and redeem government Issues and redemptions of Transactions relating to the issues and redemptions of government securities government securities securities, short-term and cash deposits Release funds to agencies Budget warrants/ Periodic release of funds by MOF to sector agencies for specific cash allocations budget categories within budgetary allocations Request adjustments to Requests for budget Request by line agency for transfer of budget allocations from one budget authorizations (may adjustments/ head to another, or for supplementary funds for specific budget involve transfer of allocations supplementary categories from one budget head to authorizations another, or supplementary authorizations to particular budget heads) Authorize budget adjustments Budget transfers/ Authorization made by competent authority (MOF/line agency supplementary official) for transfer of funds from one budget head to another, and authorizations supplementary authorizations to budget category Prepare expenditure plans Expenditure plans Line agencies' projections of expenditure, based on planned programs and projects and available budget funds Make request for goods and Procurement requests Request for procurement of goods and services made out by managers services within line agencies Authorize expenditures Expenditure/ procurement Authorized request for expenditure for goods and services or other request authorizations forms of payment. Authorization carried out by line agency managers after determining validity of request and availability of budget allocations Issue purchase orders Purchase orders Order for the purchase of goods and services issued by line agency or central supply organization specifying goods and services required and time of delivery Commit funds Transaction setting aside funds as a result of approval of specific requests for procurement of goods and services and issuance of corresponding purchase order Receive bills for goods and Bills/invoices Request for payment made by vendor to line agency for goods and services services procured by that agency against a purchase order service 168 ANNEX 12 Business Process Data Entities Definition Created Receive goods and services Goods receiving Certificate of receipt of goods/delivery of services required prior to report/certificate of release of payment completion of services Authorize payment Payment orders/ Payment Authorization for payment against a bill or invoice made by line vouchers agency finance officials or treasury/MOF officials after determining availability of funds Pay for goods and services Check Financial instrument authorizing recipient to draw money from line. agency account with the treasury or authorized servicing bank Administer payroll and Payments related to payroll Payroll payments made to civil service employees employee benefits Administer pension and Payments related to Pension payments made to government pensioners retiree benefits pension Service public debt Debt service payments Debt service payments made for government borrowings Make other payments Other payment types Payments related to grants, subsidies, etc. Receive requests for Service requests Request made by client for products/ services offered by line agency products/ services provided by line agency Invoices for goods and Invoices Request for payment made by line agency for products delivered or services services rendered Receive payments Payment receipts Financial transaction for receipt of funds (which may be for products/services delivered by agency) Receive government tax Tax revenue receipts Receipts of government tax revenues paid into the treasury revenues Receive government nontax Nontax revenue receipts Receipts of government nontax revenues paid into the treasury revenues Receive loan proceeds, Loan/grant receipts Receipts of government loan proceeds/ grants paid into the treasury grants, etc. Maintain budget and Budget ledgers Record of transactions showing amount of budget authorizations and commitment ledgers funds allocated for programs and projects and all changes to authorizations/funds allocations as a result of budget transfers or additional fund allocations via supplementary authorizations, with the authority and dates of various changes and totals of expense and commitment transactions against budget categories Maintain accounts payable Accounts payable ledgers Record of payment and payable transactions carried out over the fiscal ledgers year Maintain receipt/receivable Accounts receivable Record of receipts/receivable transactions carried out over the year ledgers ledgers Maintain general ledgers General ledgers/journal Record of financial transactions classified according to chart of entries accounts Develop cost of programs Cost-accounting reports Record of transactions recording costs incurred against programs and 169 ANNEX 12 Business Process Data Entities Definition Created and projects projects Maintain store inventory Stores/inventory ledgers Record of transactions recording physical issues and receipts of goods ledgers in stores Monitor and evaluate budget Expenditure reviews Periodic reviews of actual expenditures, analysis of variations with budgetary estimates, and comparison of financial and physical progress; consisting of overall budget reviews and agency reviews of programs and projects Periodic reports to monitor overall flow of appropriations and inflows Fiscal reports of revenues over the course of the year, highlighting major deviations from planned budget program and suggesting corrective measures Summary of Functional Requirements for the Budgeting and Accounting Modules of a Government Fiscal Management System 1 Budget and Funds Control The budget and funds control subsystems must include: * expenditure control equivalent to funds allocation; * financial and other data for budget and program estimates, and periodic budget reports; * budget control at any nominated segment of the account code; * budget control on a monthly or biweekly basis (or as required); * control available either as a warning or to prevent a * transaction taking place, as required; * on- ine checking of commitments and expenditure against * availability of funds; * adjustment to commitment at time of posting expenditure; * on-line checking of forward commitment against limits; and * formula-driven allocation function to enable a high-level * allocation function to be split, in accordance with a user * specified formula, to lower level allocations. Purchasing The purchasing subsystem must: * provide access to a list of potential suppliers and vendor files and a means of maintaining these databases; * allow recording of details of supplier performance; Extracted from the Report to the Information Exchange Steering Committee, Department of Finance, Government of Australia, December 1990. 170 ANNEX 12 * allow entry of requisitions (proposals for expenditure); * allow entry and printing of purchase orders from any approved location by approved user; * allow credit card transactions to be recorded, and provide facilities to reconcile transactions with monthly credit card statement; * allow splitting of orders to multiple accounts; * provide recording of receipt of goods/services, including partial delivery; * accommodate variations to, and cancellation of, purchase orders; and * accommodate multiple vendor addresses (e.g., different addresses for placement of order and payment). Accounts Payable The accounts payable subsystem must: * provide facilities for registration and monitoring of claims; * match purchase order, invoice, and delivery details; * process credit notes from suppliers; * generate recurring payments; * provide a due date facility; * allow payment of claims by credit card; * allow recording of details of cash payments; * accommodate many-to-many relationships between invoices and purchase orders; * notify creditors on remittance advice if discount taken; * provide facilities to enter check-butt/remittance advice details; * allow splitting of payments to multiple accounts; * record commitment if invoice does not relate to an existing purchase order; * allow purchase order to be reopened by authorized staff after final payment has been made; * verify transaction to avoid payment of duplicate invoices; and * provide techniques to handle voided checks. Accounts Receivable The accounts receivable subsystem must provide facilities for: * processing sales orders, invoices, and debit advice notes; and following up debt, including calculation of interest; * producing age and trend analyses of debts; * recording bad debts, write-offs and recoveries; * recording details of agreements with debtors; * registering credit notes from suppliers and issuing credit notes to clients; * producing bank deposit slips; * acquitting receipts against invoices; and * maintaining debtor details. General Ledger and Chart of Accounts The general ledger and chart of accounts subsystem must provide: * flexible coding system; * automatic posting to subsidiary ledgers; 171 ANNEX 12 * transaction details for previous years; * totals for previous years; * cash commitment and accrual accounting; * user-defined accounting periods; * journal entry facilities; * entry and reporting of nonfinancial information; * account maintenance facilities; * procedures associated with end of accounting periods; * facilities for recurring accrual entries; * facilities for a single transaction to have effect in two accounting periods; * mechanism for cost allocation; and * separate recording of expenditure and revenue (not just cash balance). Internal Controls and Audit The internal controls and audit system must allow the user/auditor to: * ensure that all processing complies with statutory requirements; * produce exception reports; * reconcile subsystems with general ledger; * obtain audit trails for all transactions associated with supplier and other master files; * obtain transaction details to support account balances; and * obtain audit trails to enable source documents to be traced to financial statements and vice versa. Security The security system must: * provide facilities for electronic signatures, for internal user identification and authorization, and for external authorization and check production; * log all transactions by user and terminal; * record users' identification as part of the transaction record; * limit access to data files and programs both through the system and through access methods external to the system; * prevent alteration of financial data except through the posting of transactions that are entered through the normal edit and update process; * provide a system-wide security facility covering all core modules; * allow detection, reporting, and investigation of unauthorized access to data; * prevent malicious or accidental destruction/misuse of data; * provide a security management system that allows individual and generic user security profiles to be created, and that maintains and controls access to data and functions on the basis of these profiles; * require a single user identification password to access all core modules to which access is allowed; * allow a limit to be placed on the number of unsuccessful/unauthorized attempts at a particular operation; and * provide an audit trail of unsuccessful/unauthorized attempts to access the system. Backup and Recovery The backup and recovery system must work in conjunction with the operating system, the transaction processing system, and the database management system to assist in: * identifying data files that have been changed and those that will be saved for recovery purposes; and 172 ANNEX 12 * backing up all incomplete transactions, restoring the system to its last consistent state, and reapplying the transactions that have not been successfully posted since the last backup. * This system should provide options for selective or full restores, incremental or complete backups, and on-line backups by excluding user access. 173 ANNEX 13 ANNEX 13: THE UNITED STATES: FEDERAL HIGHWAY TRUST FUND The United States established the Highway Trust Fund in 1956 to finance the federal share of the inter-state highway network and support most other Federal-aid highway projects. Later amendments extended funding to other transport programs as follows: * The Highway Safety Act of 1966 made funds available for state and community road safety programs. * In 1982, the scope was widened to permit the financing of mass transit. * In 1991 the Intermodal Surface Transportation Efficiency Act confirmed the new role of the Highway Trust Fund as an "Intermodal Fund" by extending support to high speed rail and bike-trails. The funding system involved earmarking certain road-related taxes and depositing them into a special account, or road fund. The special account was primarily introduced to finance construction of the interstate highway network and was based on the user-pay concept. The concept involves two elements: first, the user pays, and second, the government credits the user fees directly to a highway special account. The user-pay concept is well established in the United States. All but six states and the District of Columbia now dedicate their user-fee revenues to special highway or transportation accounts. The US Federal Highway Trust Fund exists only as an accounting mechanism. The taxes earmarked for the Trust Fund are deposited into the general fund of the US Treasury and a paper transfer of these taxes is made to the Trust Fund as needed. Earmarked tax revenues in excess of those required to meet current expenditures are invested in public debt and interest earned is credited to the Trust Fund. The Trust Fund finances the federal-aid highway program administered by the Federal Highway Administration. Since 1982 a portion of the Fund has also been used to finance mass transit projects administered by the Urban Mass Transportation Administration. Revenues from the highway portion of the Trust Fund are used to reimburse states for expenditures on approved projects. These include heavy maintenance (reconstruction, rehabilitation and resurfacing), road improvement, new construction, road safety programs, studies, and other highway related expenditures. The Trust Fund does not currently finance regular maintenance. Trust Fund revenues are derived from a variety of highway user taxes, including: (i) motor fuel taxes on gasoline, diesel, and gasohol; (ii) a graduated tax on tires weighing 40 lbs. or more; (iii) a retail tax on selected new trucks and trailers; (iv) a heavy-vehicle use tax on all trucks with a gross vehicle weight (GVW) over 55,000 lbs; and (v) interest on the Trust Fund balance. Tax rates are adjusted as part of the regular budgetary process. In 1995, the tax rates were: gasoline = 12 cents/gallon; diesel = 18 cents/gallon; special fuels = 12 cents/gallon; tires = sliding incremental scale which varies from 15 cents/lb to 50 cents/lb over 90 lb; a 12 percent tax on the retail price of trucks over 33,000 lb GVW and trailers over 26,000 lb GVW; $100 plus $22 for each 1,000 lb over 55,000 lb GVW up to a flat fee of $550 for trucks over 75,000 lb GVW; 174 ANNEX 13 and interest at about 6.75 percent. In 1995 the revenues from the above tax rates were $17.323 billion (about 2/3rds from gasoline), $395 million, $2.0 billion, $682 million, and $548 million respectively, to give a total of $20.967 billion for the year. An additional $2.8 billion was likewise paid into the Mass Transit Account during the year. Some vehicles like school buses and state, local government and non-profit vehicles are exempted from paying Federal Highway motor fuel taxes. In addition, fuels used in off-highway uses are exempt from these taxes (e.g., agriculture and industry. Coloring un-taxed diesel and testing non-exempt diesel vehicles to ensure they are using regular (taxed) fuel deal with off- highway uses. The federal-aid highway program is a reimbursable program. What are allocated to the states are not cash, but a line of credit against which they can draw to meet obligations. Funds are allocated on the basis of formulas that, though not perfect, are difficult to change. The US Government Accounting Office has recently criticized the formulas, but concluded "Because the selection of a highway apportionment formula is a judgment for the Congress, GAO is making no specific recommendations." In other words, the allocation formulas are (at least in the US) highly political. The formulas are relatively simple and generally use variables like population, road mileage, and traffic density. For example, maintenance funds are allocated according to the following formula: (interstate lane miles/total interstate miles)*0.55 + (vehicle miles on interstate roads/total interstate vehicle miles)*0.45 This means that the average allocation per state is about 2 percent of total maintenance allocations, subject to each state receiving a minimum allocation of 0.5 percent. Allocations do not cover all costs, but generally cover 80 percent of costs or, in the case of maintenance, 90 percent of costs (i.e., funds from the Highway Trust Fund are provided on a cost-share basis). In looking at this allocation formula, the GAO suggested consideration of two alternatives: Alternative 1 Distribution based equally on (i) total lane miles and (ii) total vehicle miles traveled. Alternative 2 Distribution based equally on: (i) total lane miles, (ii) interstate vehicle miles traveled and (iii) state population. Payment for work financed through the Highway Trust Fund is made in the following way: (i) A contractor does work (ii) Contractor is paid by the state. (iii) Vouchers for reimbursement (usually covering several project withdrawals) are sent to FHWA for review and approval. (iv) Claims are certified by FHWA (this is a formality and certification is automatic). (v) Certified schedules are submitted to Treasury. (vi) Electronic funds transfer Federal share to State bank accounts. 175 ANNEX 13 Each state participating in the scheme is required by law to have an annual audit carried out. The audits are normally carried out by outside auditors and cover financial and compliance and internal control procedures (i.e., it is more extensive that a purely financial audit, in that it also covers control procedures, etc.). Staff from FHWA also check these procedures on an ad hoc basis. There is no formal technical audit. Staff from FHWA used to carry out field inspections but this has ceased due to staff shortages. However, occasional field inspections are still carried out. FHWA is also subjected to an annual audit to ensure it follows laid down procedures and can account for funds spent. About 3,000 staff manages the federal-aid highway program. They are stationed in Washington and in each of the states. 176 ANNEX 14 Annex 14: Evaluation Capacity Development in the Australian Government Executive Summary The Australian federal government has given a high priority to ensuring that evaluations of its programs are conducted and the findings used. The approach, which has been adopted to develop evaluation capacity, has entailed a combination of formal requirements for evaluation plus their strong advocacy by a powerful, central department (the Department of Finance (DoF)). This has enabled evaluation to be linked both to budget decision-making and to the management of government programs. As a result, there is clear evidence of a high level of evaluative activity and that evaluations are actually used to assist Cabinet's decision-making and prioritization in the budget and to support internal program management within line departments. More recently, the Australian system has evolved from one of tight, formal controls and requirements to more of a voluntary, principles-based approach. In this new environment, it is hoped that the strong pressures for line departments to achieve and to demonstrate high levels of performance will be increased and that the existing evaluative culture and infrastructure will be strengthened. In particular, the latest wave of public sector reforms promise a closer integration of performance measurement-including evaluation-into performance management and governance more broadly. There are many lessons here for the development of evaluation capacity in developing countries. A number of these lessons are identified and discussed in this Annex. Genesis and Stages of Development An understanding of the development of evaluation capacity in the Australian government is important because of the insights it provides for other countries. Evaluation capacity development (ECD) did not develop in a linear, logical sequence in Australia; rather, it grew in an opportunistic manner in response to prevailing imperatives. The twists and turns of these developments mean that a large body of experience has been amassed concerning what works, what does not, and why. The objective of this annex is to share these insights. As with other developed countries, evaluation is not a new phenomenon in Australia. Cost-benefit analysis-the appraisal of investment projects has-been part of the scene for many decades. It has been conducted by the various government economic research bureaus that have existed, and by other advisory bodies and line departments concerned with infrastructure and agriculture investments. Formal program evaluation has a more recent origin. It too has been conducted by specialist research bureaus attached to line departments, and also by special policy review 177 ANNEX 14 taskforces, focusing on areas such as labor market programs and social welfare policy. But as a discrete and on-going focus of government activity its heyday started in the 1980s. The election of a reformist Labor government in 1983 provided an environment that was strongly encouraging of evaluation. The new government was determined to improve the performance of the public sector while at the same time achieving tight public expenditure restraint through the annual budgetary process. A series of public sector management reforms was implemented in the first several years of the new government. One aspect of these reforms was the desire to 'let the managers manage' by devolution of powers and responsibilities-reflecting the philosophy that public sector managers would be strongly encouraged to improve their performance if they were provided greater autonomy and the potential to manage their departments with fewer central agency controls and less interference. Another aspect of the reforms was 'making the managers manage', and this thread was more directly linked to the powerful budgetary pressures which existed. The more tangible changes included: * substantial autonomy for departments in their spending of administrative expenses (including salaries), but with these administrative expenses being strictly cash-limited; * greater surety about future resource availability to departmental managers via a system of three-year forward estimates of administrative and other program expenses; and * a major reduction in the number of departments through amalgamation, in order to achieve less balkanized policy advice and to encourage the internal reallocation of resources through portfolio budgeting. A related set of principles was embodied in the Financial Management Improvement Program, which included program management and budgeting. These principles emphasized the importance for departmental managers in ensuring that their program objectives were realistic and to help guide managers and staff alike. The principles also encompassed a focus on the efficiency and effectiveness of programs-i.e. on program performance-through sound management practices, the collection of performance information and the regular undertaking of program evaluation. Guidance material extolling the virtues of these principles and clarifying concepts was issued by the DoF together with the then Public Service Board, another central agency. These initiatives were introduced in a stringent budgetary climate. Macroeconomic concerns provided sufficient motivation to induce the government to reduce the share of federal government outlays in GDP from 30 percent in 1984-85 to just over 23 percent in 1989-90. The impact of these cuts on government programs was even greater than these raw statistics would suggest, because the government was also determined to increase the real level of welfare payments to the most disadvantaged in society. It achieved this by means of tight targeting of benefits, and this entailed substantial reductions in 'middle-class welfare'. This package of financial management and budgetary reforms was substantive and wide- ranging. For several years, it placed the framework of public sector management in Australia at the forefront of developed nations. 178 ANNEX 14 DoF was a major architect of many of these reforms, reflecting its role as budget coordinator and overseer of the spending of other departments. DoF was keen to get out of the detail of spending issues, where it was often bogged down in minor spending bids and disputes with departments. Its concern with budget spending encompassed not simply a priority on cutting government outlays, but also in finding ways to make spending more efficient and effective. This concern with value for money helped to lay the groundwork for DoF's provision of advice to departments on the evaluation of their programs. It commenced this role in a formal sense with the publication of an evaluation handbook in 1986. At around the same time there was growing disquiet in DoF and other central agencies about the lack of real progress made by line departments in managing their performance. In early 1987, the Minister for Finance took the lead role in securing Cabinet's agreement to a formal requirement that all new policy proposals for Cabinet consideration should include a statement of performance measures as well as proposed arrangements for future evaluation. Line Ministers and their departments were also required to develop plans for the systematic and comprehensive monitoring and evaluation of the performance of their programs, and for the reporting of these reviews to government. DoF was also to be kept informed of departments' evaluation plans. DoF augmented its earlier advisory role through the provision of additional guidance material in 1987, and via some limited provision of a basic evaluation training course. By early 1988, however, the evaluation plans prepared by departments could best be described as patchy, and many were poor, and it had become apparent that a more fundamental examination of evaluation practices in departments was warranted. Thus DoF undertook a diagnostic study reviewing departments' evaluation progress and the overall health of evaluation activities in the public service. The study found: * a lack of integration of evaluation into corporate and financial decision-making; * that evaluations tended to focus on efficiency and process issues rather than on the more fundamental question of overall program effectiveness; * a poor level of evaluation skills and analytical capacity; and * that the role of central departments, especially DoF, was unclear. This diagnostic study laid the groundwork for a major submission from the Minister for Finance to Cabinet in late 1988 seeking-and securing-its agreement to a formal, ongoing evaluation strategy for all departments. A key, and continuing, principle underlying this strategy was that 'the primary responsibility for determining evaluation priorities, preparation of evaluation plans and conduct of evaluations rests' (with departments). The evaluation strategy had three main objectives. It encouraged program managers within departments to use evaluation for the improvement of their programs' performance. It also provided fundamental information about program performance to aid Cabinet's decision- making and prioritization, particularly in the annual budget process when a large number of competing proposals are advocated by individual Ministers. Lastly, the strategy aimed to strengthen accountability in a devolved environment by providing formal evidence of program managers' oversight and management of program resources. The emphasis was on transparency, 179 ANNEX 14 which is of interest to the parliament, particularly in the senate's processes of budget scrutiny and approval. The evaluation strategy has provided the framework and driving force underlying the progress with (ECD) in the Australian government since that time. Its key components comprised four formal requirements. These were: * that every program be evaluated every 3-5 years; * that each portfolio (i.e. comprising a line department plus outrider agencies) prepare an annual portfolio evaluation plan (PEP), with a 3-year forward coverage, and submit it to DoF. PEPs comprise major program evaluations with substantial resource or policy implications; * that Ministers' new policy proposals include a statement of proposed arrangements for future evaluation; and * that completed evaluation reports should normally be published, unless there exist important policy sensitivity, national security or commercial-in-confidence considerations, and that the budget documentation which departments table in parliament each year should also report major evaluation findings. A crucial aspect of this evaluation strategy was the role that Cabinet agreed to assign to DoF. Cabinet expressed its expectation that DoF would have the opportunity to make an input to PEPs and to the terms of reference of individual evaluations to ensure that they were consistent with government-wide policies and priorities, and that DoF would be available to participate directly in selected evaluations, subject to negotiation between DoF and the line department (or between their Ministers if a dispute arose). DoF was also assigned a role in providing detailed advice and handbooks on evaluation methodology, and on management information systems, and in taking the lead in identifying and sharing best practices. A key to the public sector reforms that DoF had advocated was the need for central departments to get out of the detail and out of a control mentality. While the leadership of DoF were true believers in the utility of evaluation, it could be argued that they had advocated a more hands-on involvement for DoF with some considerable hesitation, because it represented an activist and closely monitoring approach. DoF's advocacy of the powerful role that it was assigned by Cabinet only came after there was clear and strong evidence of the failure of line departments to live up to the rhetoric of program management and budgeting which they also had espoused. While the evaluation strategy was still at the proposal stage, most line departments stated their acceptance of the utility of evaluation but expressed real concern about 'intrusive DoF involvement' This stated acceptance of the principles was at odds with the actual reality of management in most line departments, and was a feature of other performance measurement and performance management initiatives introduced in later years. That Cabinet agreed to DoF's role in the face of sustained opposition from line departments almost certainly reflected the influence of the Minister for Finance, and of his department, in a period of very tight expenditure restraint. Evaluation offered the promise of a tool to help better inform Cabinet's decision-making and prioritization among different spending and savings options. 180 ANNEX 14 The next milestones for ECD in Australia were two reports, from a parliamentary committee in 1990 and from the national audit office in 1991. These reports acknowledged the substantial effort being devoted to the planning and conduct of evaluation, but argued for renewed efforts to be made. In particular, they noted with some concern the variation in the extent of evaluation activity in different departments, and argued for a more active role by DoF to encourage evaluation. These reports were soon followed by the creation in 1991 of a separate branch within DoF, responsible for the provision of evaluation advice, support, training and encouragement to other departments and also within DoF itself. This branch had 9 evaluators able to provide assistance and it acted as a focal point and catalyst for evaluation throughout the Australian public service. The branch was also co-located with two other branches responsible for overall coordination and management of the government's annual budget process and for public sector management reforms more generally. As discussed in the following section of this annex, evaluation in the Australian government-as measured by the extent of evaluation planning, conduct and use-had achieved a healthy and vigorous state by the mid-1990s. However, by that time DoF was concerned about departments' poor progress in articulating clear and achievable objectives for their programs and in collecting and reporting meaningful performance information on a frequent basis. These concerns were confirmed by two reviews of departments' annual reports and of their budget documentation, which DoF commissioned. This evidence was somewhat ironic, because evaluation is at the difficult end of the performance measurement spectrum, and it was generally being done well, yet program objectives and performance information are at the easier end of the spectrum, and they were being done poorly. Thus in 1995 DoF secured Cabinet's agreement to a rolling series of comprehensive reviews, staggered over 3 years, of the program objectives and performance information of all departments. These reviews are being conducted jointly by DoF and each line department, with the results being reported to their respective Ministers and to Cabinet as a whole. The reviews focus in part on the existing state of play, but more importance is attached to identifying ways in which objectives and performance information could be improved in the future, and in mapping out and committing to a plan of action to achieve these improvements. This perhaps illustrates, once again, that public sector management initiatives often require a push from the center to make them happen in many departments. The election of a Liberal/National Party government in 1996 has led to an emphasis on cutting bureaucracy, red tape and the large number of formal reporting requirements. The new government has required that existing programs be comprehensively reviewed by Ministers and their departments to establish whether or not they should continue, or whether they should be abolished or devolved to another level of government. For those programs which are to continue to be delivered at the federal level, there is an expectation by Cabinet that a competitive tendering and contracting process be undertaken wherever possible. 181 ANNEX 14 One concern that has emerged with the formal evaluation requirements in recent years is that their concern with bureaucratic process is no longer appropriate-thus the length of some portfolio evaluation plans, for example, has grown from a recommended 20 or 30 pages in length to over 120 pages. A consensus has emerged within the bureaucracy that while it is important to have evaluation findings available to assist decision-making by program managers and by Cabinet, detailed and elegantly-worded plans are not necessary to achieve that end. There is also a related and powerful strand of thinking that departments should not be encumbered by excessive controls on their internal activities, as long as departmental heads and senior executives are responsible for performance, and that responsibility is reflected in their employment contracts. This is essentially a 'let the managers manage' philosophy, and is analogous to the one adopted in the early 1980s. A difference, however, is the greater scrutiny by Ministers of the performance of their department heads-as exemplified in the more widespread use of employment contracts-plus the apparent readiness of Ministers to remove department heads from their positions where their performance is judged to be wanting. And perhaps the most important difference of all is the progress in the intervening years in establishing an evaluation culture, and in establishing departmental infrastructures to support performance measurement and management. This evolution in philosophy led, in late 1997, to a further development in the evaluation strategy into a principles-based, performance measurement framework (i.e. evaluation plus performance information). Such an approach has been accepted by Cabinet, and is now government policy. Such a Cabinet-endorsed, principles-based approach provides guidance to the heads of line departments by emphasizing the good practice features of performance management and measurement. It implicitly reflects the strong expectation that CEOs and senior executives will continue to plan, conduct and use evaluation, and so it implicitly takes the progress achieved to date as a given. These issues go to the heart of the question whether central controls should be tight or loose. Evidence of Progress with ECD It is important to have a realistic understanding of the extent of ECD in Australia. All too often, national reviews of progress with evaluation in developed countries present too rosy a picture, particularly when external reviewers confuse rhetoric with reality. Checklists of evaluation activities undertaken do not necessarily translate into the widespread conduct of evaluation, nor into quality and rigor in evaluation, nor into its actual use in program management and government decision-making. The extent of ECD in Australia can be assessed by considering the planning, conduct, quality and use of evaluation. Evaluation Planning All government departments prepare portfolio evaluation plans, and have done so since 1987-1988. These are intended to comprise the major evaluations in each department and its outrider agencies-in recent years about 160 of these evaluations have been underway at any given time. The majority of these evaluations are major, in that the programs have significant 182 ANNEX 14 policy or spending implications, although a significant minority of these evaluations, particularly for the smaller departments, is only of minor programs or of efficiency aspects of large programs. (The plan guidelines issued by DoF recommend that the main focus of these evaluations be on issues of program effectiveness. Departments are separately encouraged to plan and to undertake more minor evaluations for their own internal management purposes.) Line departments themselves decide which programs should be included in the plans for evaluation, and also which issues the evaluation terms of reference would cover. However, DoF would usually endeavor to influence departments' choice of evaluation priorities by making direct suggestions to them. In making these suggestions, DoF would attempt both to anticipate and to create the information needs of Cabinet. Where DoF has had difficulty in persuading departments, it has sometimes approached Cabinet directly to seek its endorsement of particular evaluation suggestions and also detailed terms of reference; Cabinet almost invariably accepts DoF's suggestions. The Cabinet-endorsed, formal requirement under the evaluation strategy that portfolio evaluation plans be prepared and submitted to DoF certainly provided a powerful incentive to line departments to prepare plans and to take them seriously. Another influential factor was DoF's issuance of formal guidelines to departments on the desirable content of these plans, together with follow-up monitoring and reminders to departments about the need for the plans. The evaluation branch of DoF conducted internal reviews of the content and coverage of these evaluation plans and provided feedback and prompting to departments as well as identifying good practice examples. More formal feedback and analysis has also been provided by the national audit office on several occasions, via several efficiency audits into program evaluation which it has conducted. The formal requirement that all programs be evaluated every 3-5 years was also influential in creating a climate of expectation that evaluation is the norm rather than the exception. The concept of regular, comprehensive coverage of programs also encouraged a planned, staged approach to evaluation. This formal requirement should not be accepted at face value, however. It is very seldom the case that all aspects of a program' are included in any single evaluation. Instead, it is usual that an evaluation will focus only on certain key problems or aspects of a program. The challenge is to ensure that these difficult issues are actually evaluated, and this is a role in which DoF has played an active role via persuasion and via direct involvement in individual evaluations., The three dimensions on which program evaluation can focus are (i) the efficiency of a program's operations, (ii) its effectiveness in achieving its objectives, and (iii) whether the program's objectives remain consistent with the government's policy priorities, i.e. the appropriateness of the program. 2 There was only modest success with the requirement that Ministers' new policy proposals include an Avaluation plan of action that would be undertaken if the proposal was accepted. Feedback from portfolios indicated that this requirement was onerous for portfolio managers during the busy budget period. Only about 30 percent of proposals broadly met this requirement in the 1993- 1994 budget, for example, although an additional 50 percent of proposals included a clear undertaking to evaluate the proposal if accepted (DoF 1994). These percentages were only achieved after considerable prodding by line areas within DoF. In recent years the extent of such prodding and of departments' willingness to provide such plans in their budget documentation has fallen off considerably. 183 ANNEX 14 Conduct of Evaluation The large number of evaluations in progress, and the fact that over 530 evaluation reports have been published over the last four years or so, attest to the existence of extensive evaluation activity in the Australian government. This has provided a growing 'library' of evaluation findings. DoF maintains a register of published evaluation reports, and this helps to monitor the progress of individual departments' activities. More importantly, it helps to share evaluation practices and methods among departments and this also provides some quality assurance because the public availability of these reports exposes them to peer scrutiny.' Evaluation Quality Quality of evaluation reports is a more difficult dimension to measure. The rigor of program evaluations depends on the expertise and objectivity of the evaluators. The national audit office commissioned a recent assessment of the quality of a small sample of evaluation reports. It found that over a third of a sample of evaluation reports suffered from methodological weaknesses of one kind or another. It is certainly the case that some published evaluations are of low quality, and the suspicion is that some of these are produced for self-serving purposes, such as to provide a justification for the retention or expansion of the program. DoF's own experience of evaluations is that their quality can vary enormously. The extent to which this should be a matter of concern is another matter, and an issue to consider here is the intended uses of evaluations. If the intended audience of an evaluation is Cabinet (to aid its decision-making) or the parliament (for accountability purposes) then a poor quality or misleading evaluation provides particular cause for concern. DoF has certainly been willing to provide Cabinet with a dissenting view on the quality of an evaluation in those cases where an evaluation is used by a line department to attempt to influence a Cabinet debate. (Line departments would typically try hard to avoid such disagreements, which would be virtually guaranteed to attract the ire and condemnation of Cabinet.) Where line departments choose or allow poor-quality evaluations to be conducted and these evaluations are intended for internal program management purposes, however, then there is perhaps an element of caveat emptor. Thus the extent of evaluation quality assurance or quality control could be regarded as an issue for departments' internal management to address. A commonly asked question is how evaluation quality can be ensured and what the role of DoF is or should be in guaranteeing quality. In past years, parliamentary committees and the national audit office have argued that DoF should take a strong role. DoF has not assumed a formal role here, however. It has preferred to seek to participate directly in certain major evaluations, usually via steering committee membership, thus ensuring that evaluations address the difficult questions and do so in a rigorous manner. But it would be a very resource-intensive activity to undertake detailed reviews of all evaluations, and it would also be inconsistent with the devolutionary reform philosophy for it to assume such a role. A survey of all departments conducted by the national audit office found that 75 percent of evaluations conducted in 1995 and 1996 were released to the public and/or available on request (Australian National Audit Office (ANAO) 1997). 184 ANNEX 14 The national audit office has consistently argued that departments should set up central oversight procedures to achieve quality assurance of evaluations conducted by line areas within the department. There is certainly evidence from those few departments that have followed this approach, that it is an effective means of bringing to bear needed evaluation skills and expertise, and of ensuring evaluation quality. Use of Evaluation A bottom-line issue is the extent to which evaluation results are actually used. If their use is patchy or poor then there really is little point in conducting evaluations. The large volume of evaluation activity in itself provides some reassurance that evaluation findings are being used- in an era of very tightly limited administrative expenses, departments would simply not bother to conduct evaluations unless they were going to be used. (And DoF also would not bother to advocate and work to influence the evaluation agenda unless it perceived real potential value in their findings.) There is clear evidence that evaluations have been used intensively in past years in the budget process. DoF has conducted several surveys of the extent of the influence of evaluation findings on the budget proposals submitted to Cabinet for its consideration. These have been surveys of DoF officers to seek their judgment concerning the extent of the influence of evaluation. While the survey results should be regarded as providing no more than a broad indication of the extent of influence, they are nonetheless very revealing. In the 1990-1991 budget, some A$230 million of new policy proposals submitted by line Ministers were judged to have been directly or indirectly influenced by the findings of an evaluation. By 1994-1995 - the latest year for which estimates are available-this had risen to A$2300 million. (Measured in dollar terms, the proportion of new policy proposals influenced by evaluation rose from 23 percent to 77 percent over that period.) In most cases the influence of evaluation was judged by DoF to be direct. 1 I.94 r. I% o e% aluaion acu. nics in one of the largest departments. the Department of Employment, FJu(.,non .and I rainin--, (DEE H found that considerable use i%as being nade of evaluation findings in that dcpartment i The reviil. conducted jointly bN DEET and DoF. sur%e%ed a large sample of completed c%. luaiions and tound that * pkrccnt had led t Lhan-es in program management: * 8 perL T lud rve.ulied in an impro%ement in the qualit% of program outputs. * Ilu percent had influenced new polic% proposals sent by the line Minister to Cabinet for its consideration; hut * tha thLrc %%as considerahle univenness within DFET in the qualit and use made of evaluation findings wmnie lne :irca% ot DEFT had maintained a fiiah standard. N%hile other areas %%ere either consistently P1r0 L) %ei e unc% en The recent national audit office survey found that, for evaluations conducted over the period 1995 - 1997: about half examined the delivery of products or services to external clients; a further 30 percent were associated with matters internal to the department; one third of the evaluations examined the appropriateness of new or established programs; and 15 percent were directed at the development of policy advice for the government (ANAO 1997). 185 ANNEX 14 There were some unique features of the 1994-1995 budget that resulted in these figures being particularly high. Nevertheless, the results indicate the importance which public servants- in their preparation of the details of new policy proposals-and Ministers have attached to having evaluation findings available. Overall, it has been very important to have the active support of key Cabinet and other Ministers in encouraging portfolios to plan and conduct high-quality evaluation. It is also the case that evaluation can have a significant influence on the savings options put forward by DoF or by portfolios for Cabinet consideration in the budget process. In 1994- 1995 about A$500 million of savings options-or 65 percent of the total- were influenced by evaluation findings. One issue that is important to appreciate is the realistic limits to the influence of evaluation on Ministers' or Cabinet's decision-making. The evaluation paradigm in an investment project is typically that of cost-benefit analysis: a project is warranted if, but only if, its benefit-cost ratio is greater that one. But program evaluation is a more qualitative science: it can help identify the efficiency or effectiveness of existing, ongoing programs but it can rarely provide an overall conclusion that the activity is worthwhile or not. To give an example, if government decides to allocate a large amount of spending to the unemployed, program evaluation findings can help to map out the likely consequences of alternative types of labor market intervention such as wage subsidies, public sector job creation, labor market regulation, etc. In other words, program evaluation can be used to attempt to identify the cost-effectiveness of each type of intervention. But program evaluation can usually not identify the overall amount of resources that should be allocated. That is a political decision, which is why governments are elected in the first place. Thus the most that program evaluators can realistically and legitimately hope for is that their findings are an influential input into government's decision-making and prioritization among competing proposals and programs (see box on previous page).5 Success Factors - What Has/Has Not Worked, And Why The preceding discussion of the development of ECD and of progress achieved to date has identified a number of success factors. These are considered below, together with other success factors and some thoughts on their relative importance. The Department of Finance DoF has played a leading and central role in ECD in the Australian government, and there have been advantages and some disadvantages in this. Overall, however, DoF's considerable clout with departments and with Cabinet has given it a powerful and influential role in promoting evaluation. This has been perhaps the single most important factor in the substantial degree of progress with ECD in Australia. In the absence of evaluation findings, decisions will likely be influenced more by ex ante analysis and/or anecdotal information and case study examples. 186 ANNEX 14 DoF has an influential, devil's advocate role in advising Cabinet about the level of funding which should be allocated to departments for different government programs. As part of this function, it provides advice on the new policies proposed by line Ministers and on possible 'savings options', i.e. areas of government expenditure which could be trimmed or abolished entirely. This role does not endear it to other departments, and in fact is an impediment to close cooperation and trust between other departments and DoF. On the other hand, DoF also plays a day-to-day role with departments in advising on funding and policy issues, participating in reviews and evaluations of their programs, and providing advice on evaluation and other public sector management tools. In the evaluation area, DoF's evaluation branch, with its nine evaluators, provided desk officer assistance to departments with advice on methodology, best practice, provision of training courses, publication of numerous evaluation handbooks and guidance material, and support for evaluation networks of practitioners. The nature of these relationships can vary considerably, with other departments viewing DoF at best as a useful source of advice and treating it with wary respect, and at worst with downright hostility. The former relationship is much more common than the latter, however. As the evaluation 'champion', DoF has succeeded in getting evaluation 'on the agenda' in its work with departments. This perhaps stands in contrast to the situation which faced the former Office of the Comptroller-General (OCG) in Canada in the early 1990s. OCG was a stand-alone, specialist body also responsible for attempting to influence line ministries to adopt evaluation as a management tool. The difficulty faced by OCG was the perception that it was tangential to mainstream government activities, and this undercut its influence. It is interesting to note that OCG's functions have now been relocated as part of the Treasury Board Secretariat, the Canadian equivalent of DoF, in order to increase its clout in dealing with line ministries. This relocation was undertaken after a review in 1993 by the Canadian Auditor General of overseas practices, including in particular Australia' S.6 Another advantage of having DoF responsible for evaluation oversight is that it ensures a direct influence on the line areas of DoF, which oversee the line departments. Before the devolutionary reforms of the past fifteen years, DoF was heavily involved-some would even use the phrase 'bogged down'-in the detailed scrutiny of departments' spending activities. The more recent focus on evaluation and other public sector reforms has helped foster a greater focus in these line areas on bottom-line outcomes and value for money; DoF is simply too important a bureaucratic player to allow it to remain with outmoded attitudes and activities. However, achieving this needed cultural change in DoF has been a slow process over a number of years and has involved substantial staff turnover. The greater focus on value for money has also flowed through to the nature and quality of policy advice that DoF provides to Cabinet. That advice has increasingly drawn on available evaluation findings, thus also helping to raise the profile of evaluation with line departments. DoF's involvement in selected evaluations also provides some quality assurance to Cabinet about the evaluation findings on which proposals for new policy might be based. Auditor General of Canada 1996. 187 ANNEX 14 An Example of the Use of Evaluation to Help Government Cut and Reprioritize its Programs In the 1996-1997 Budget, the new government was determined both to reduce and to reprioritize government spending. Particular focus was given to labor market and related programs, which accounted for spending of A$3,800 million annually. The Minister for employment articulated the government's overall policy goal as providing assistance to the long- term unemployed and to those at risk of entering long-term unemployment. This focus was adopted both for equity and efficiency objectives-the latter pursued by achieving a better matching of labor supply and demand, At the same time, the minister wanted to achieve better value for money from labor market programs in the tight budgetary environment. Australian and international evaluation findings were drawn on heavily to help guide the policy choices made. The Minister highlighted the relative cost-effectiveness of different labor market programs. A key measure of this was estimated by calculating the net cost to government for each additional job placement from different programs-as measured by the increased probability of an assisted person being in a job some 6 months after they had participated in a labor market program. (The baseline was a matched comparison group of individuals who did not participate in a program.) Evaluation findings showed that the JobStart program, which provides wage subsidies, had a net cost of A$4,900 per additional job placement, whereas the JobSkills program, which was a direct job creation program, had a net cost of A$76,600. The Minister noted that "the Government will be concentrating its efforts on those programmes which have proven most cost-effective in securing real job outcomes". As a result, the JobStart program was retained while the JobSkills program was substantially scaled back and more tightly targeted to jobseekers that were particularly disadvantaged. Total savings to the government from its reduction and reprioritization of labor market programs were about A$1,500 million over two years. Cabinet also commissioned a series of major evaluations into its new labor market programs and into the new arrangements for full competition between public and private employment service providers. Source: Senator Vanstone (1996); DEETYA (1996); Commonwealth ofAustralia (1996). An interesting example of the evaluative culture which has grown in DoF, and arguably in the then Cabinet itself, was Cabinet's agreement to commission some 60 major reviews of government programs. These reviews had been suggested by DoF in the 1993-1994 to 1995- 1996 budgets, with most focusing on issues of effectiveness and appropriateness; many of these reviews surveyed and summarized existing evaluative information, rather than conducted in- depth evaluations themselves. The reviews related to aspects of programs that collectively involved about A$60 billion in annual expenditures. These reviews were designed as an urgent response to emerging budget pressures, and might best be viewed as complementary to the regular cycles of evaluation as reflected in portfolio evaluation plans. Such Cabinet-endorsed reviews can also be a useful vehicle for DoF to use if line departments strongly resist DoF suggestions about evaluation priorities. 188 ANNEX 14 A benefit to line departments from DoF involvement in individual evaluations is that it can draw on a range of evaluation skills and experience, spanning the whole breadth of government activities. Most DoF officers are usually not specialists in technical evaluation issues -they are expenditure and financial policy specialists. Perhaps their greatest potential value- added is the objective approach which they can bring to bear to an evaluation-DoF officers are comfortable in asking the difficult questions about a program's performance and in steering an evaluation towards these issues. This independent and questioning approach has provided a useful counterpoint/balance to the in-depth knowledge but often-partisan approach of program areas within line departments. Line Departments Most of the day-to-day work of line departments relates to ongoing program management. As discussed earlier, one of the three objectives of the evaluation strategy is to encourage program managers to use evaluation for the improvement of their programs' performance. This has proved surprisingly difficult at times. To those who understand the potential contribution of evaluation, its utility seems almost self-evident. Thus evaluators often plea along the lines 'how can you program managers improve your programs and better meet the needs of your clients unless you carefully evaluate program performance?' Unfortunately, motherhood appeals to the professionalism of senior executives are not terribly effective. Most managers understand the potential benefits of evaluation, but often do not view it as being at the top of their day-to-day priorities, particularly in a climate of tightly constrained and diminishing resources.7 Experience within line departments in Australia indicates that a highly supportive culture is necessary if major evaluations are to be planned, resources allocated to properly manage and undertake them, and the findings implemented. The commitment of departmental secretaries (the CEO) to achieving improvement in program performance is paramount in fostering such a culture. Over at least the past decade, the tenure of secretaries has often been brief, lasting only a few years. This turnover has meant that some departments have had a series of secretaries who have varied considerably in the priority that they have attached to evaluation and to the departmental effort that should be devoted to it. While an evaluative culture can be slow to build up, it can and has been reduced much more rapidly on occasion. The earlier discussion on the reprioritization of labor market programs provides one, high-profile example of the potential benefits of evaluation to Ministers and their departments. More generally, there has been an emphasis in recent years on 'portfolio budgeting'. This includes the setting by Cabinet of portfolio spending targets at the start of each annual budget round. In the tight budgetary environments that have predominated, the nature of these targets usually imply that if Ministers wish to propose any new policies then they must be funded from within the portfolio's existing spending envelope. Evaluation has been one tool to assist Ministers A 1992 evaluation of public sector management reform in Australia concluded that "there is widespread acceptance of the importance of evaluation". But it went on to note that "the bulk of (senior executive managers) state that it is using evaluation information only sometimes or infrequently during the conduct of their job. This suggests that information generated by evaluations is not yet a key element in program management". (Task Force on Management Improvement 1992, pp. 378 and 379.) 189 ANNEX 14 and their secretaries not only in the design of new policies but in the prioritization among existing policies (subject, of course, to Cabinet's endorsement). An advantage of the requirement for portfolio evaluation plans is that it has necessitated that departments set up a bureaucratic infrastructure to prepare them (which may be a flaw when the government is determined to cut red tape and the bureaucracy, as at present). In about three- quarters of departments this has involved the creation of a committee, usually chaired by a deputy secretary of the department, to meet regularly, to canvass candidate programs for future evaluation, and to monitor the progress of evaluations already underway., This work itself generates bureaucratic momentum. Most if not all departments also involve their Minister and their Minister's office by seeking their comments on and clearance of the draft evaluation plans. A large majority of departments have chosen to set up evaluation units to coordinate their formal evaluation planning. At their smallest, these units comprise two or three individuals. In some departments, such as employment, there is a separate branch of 20-25 staff responsible for evaluation planning, provision of advice on evaluation methodology, participation in steering committees, and the conduct of a number of major evaluations, particularly in the area of labor market programs (but typically not of education programs, which also comprise a substantial proportion of the department). It is difficult to speculate with any confidence how the evaluation 'scene' in Australia would have looked in the absence of a powerful champion such as DoF. Some of the larger departments, such as the departments of employment and health, would no doubt have had a substantive evaluation effort in any event. However, informal discussions with senior executives in those departments have emphasized the important catalytic role played by DoF even in their departments. Executives responsible for the central evaluation areas in line departments have generally found DoF to be a natural ally in helping to persuade the more traditional administrators in their departments to adopt evaluation as a valued management tool. Most departments have chosen to rely on program managers and their staff for the actual conduct of evaluations. This devolutionary approach has helped to 'mainstream' evaluation as a core activity of each line area, and has ensured that evaluations draw heavily on the program knowledge and experience of those who actually manage the program. It has also led to a greater appreciation of the complementarity-and sometimes the substitutability-between in-depth program evaluation and the more frequent collection of ongoing performance information. The devolutionary approach has also been important in securing the 'ownership' by program managers of the evaluation findings. These are important advantages, and provide a strong contrast with externally-conducted evaluations, reviews or performance audits where lack of program knowledge and commitment to implement the findings has often significantly undermined their impact. But there have also been disadvantages to this devolved approach. One has been a lack of evaluation skills in many program areas and inexperience in conducting evaluations, as suggested by the recent survey by the national audit office of a sample of evaluation reports. 'See ANAO 1997. 190 ANNEX 14 Basic training in evaluation skills is widely available in the Australian government- provided by DoF in particular'-as is DoF and departments' own guidance material such as evaluation handbooks. There is also a fairly large community of evaluation consultants in Canberra. However, the recent national audit office study also revealed that 20 percent of departments are concerned about the lack of available training in advanced evaluation techniques. Some departments have addressed the need for more advanced skills and experience by setting up a central evaluation unit to provide advice on methodology and to participate in evaluation steering committees. The department of health has pursued evaluation quality assurance in a devolved environment via ensuring adequate skills and resources are available to program managers together with having structural arrangements in place, such as technical panels and steering committees.! Another disadvantage to the devolved approach is that program staff are often too close to their program to view it objectively and to ask the hard, fundamental questions concerning its performance and the need for the program to continue to exist in the future. Again, external participation by a central evaluation unit or by peers from other programs in working groups or steering committees have been one way of attempting to address this. External participation has often included DoF for major evaluations, and this has been another means of fostering objectivity and rigor. In only one agency, the Aboriginal and Torres Strait Islanders Commission (ATSIC), is there a separate evaluation body-the Office of Evaluation and Audit (OEA)-that has statutory independence. The independence of the OEA helps to provide an answer to claims of ethical difficulties and corruption in the administration of some ATSIC programs. A body such as OEA can be effective in ensuring that accountability objectives are met. The impact of its evaluations and reviews can be reduced, however, if they are perceived to lack fluency in program understanding and if they do not secure ownership by ATSIC program managers. Australian National Audit Office The national audit office has played a valuable role since the inception of the formal evaluation strategy. In endorsing the strategy, Cabinet agreed with the proposition that 'it is expected that (the national audit office) would contribute to the proposed evaluation strategy through audits of evaluation processes within departments and agencies, including their follow- up on evaluation findings . Since 1990, the audit office has pursued this 'sheepdog' role vigorously, both with respect to line departments and to DoF. (It is notable that the Canadian Auditor General has played a similar role.)" The Australian National Audit Office (ANAO) has conducted six performance audits into the evaluation and performance information practices of a number of departments during that period, as well as on the overall, government-wide state of play with the DoF has provided introductory evaluation training to over 3000 public servants since 1991. The department of health encourages quality evaluations through: selection of good quality officers to manage the evaluation; involvement of internal and external stakeholders; ensuring technical advisory panels are available to help assess the work of consultants; having steering groups available to help manage consultants; and ensuring sufficient resources are available for the evaluation. ' See, for example, Auditor General of Canada 1996. 191 ANNEX 14 evaluation strategy. It has also published two good practice guides, one of which was prepared jointly with DoF. In addition to these reviews of evaluation and performance information, the national audit office has also placed an increasing emphasis on the conduct of performance audits into the economy, efficiency and effectiveness of programs. The audit office completes about 40 performance audits annually, and they now account for about one half of the office's overall activity. The audit office takes care to ensure that its performance audit activities - which can be regarded as a form of evaluation - do not overlap or duplicate those of departments, and so do departments and DoF when planning their own evaluation priorities.,, The impact of the audit office's activities has been felt in several ways. First, they have focused attention on evaluation as a legitimate and important area for senior management attention in departments. A second impact was felt in the earlier years, when the audit office pursued its performance audits into evaluation (and into program administration more generally) with a 'gotcha', faultfinding zeal. The impact and value-added of such an approach is highly doubtful as it strongly discouraged the 'victim' departments from ownership of the audit findings. The resistance of these departments to accepting the audit office findings was often evident in their formal, published responses to the audit office reports. A 'gotcha' approach may have satisfied a narrow interpretation of the accountability function of an audit office, particularly in its reporting to parliament, but it undermined the potential value-added contribution that a considered performance audit could provide to a line department. In more recent years, with a new Auditor-General and a different audit philosophy in the office, there has been a stronger emphasis on finding ways to help departments improve their performance. A high priority has also been attached to the identification and sharing of good practices, and the audit office has been pro-active in disseminating these among departments. Other Success Factors Having a government-wide evaluation effort, involving all departments, has proved helpful in developing a general climate of expectation that evaluation will be conducted and used, and in developing an evaluation community, especially in Canberra. It has also helped to develop a labor market for evaluation skills, including advanced data analysis skills. The labor market includes the growing number of staff with experience in evaluation units, in the various economic research bureaus, and in the national statistical agency. One expression of this community has been the monthly meetings of the Canberra Evaluation Forum. The meetings have been organized by a steering group of departments, with DoF support, and each involves several speakers and discussants of topical evaluation issues. About 100 participants attend each month. 2 Auditor-General of Canada 1996; ANAO 1997b. 192 ANNEX 14 The sharing of insights and good practices through these meetings has strongly encouraged networking. There have also been several special-interest conferences and seminars on particular evaluation issues organized by DoF and others, on issues such as the evaluation of policy advice and evaluation/audit links. A feature of the Australian scene has been the frequent availability of commercially- organized, for-profit, conferences on evaluation and other performance measurement issues, and on public sector reform issues more broadly. Various departments, including DoF, work collaboratively with conference organizers to identify topical issues and provide conference speakers. The conferences allow an opportunity for federal public servants to be exposed to evaluation issues in state governments, local government and the private sector, and academia. However, the contribution of academia to ECD in the federal government has been limited. The role of parliament has not lived up to the somewhat ambitious aims of the designers of the evaluation strategy, who viewed accountability to the parliament as one of the foundations of the public sector reforms, including the evaluation strategy. In practice, parliament has generally possessed neither the infrastructure resources nor the perspective to focus on the insights into program performance which evaluation findings can offer. While parliament exercises a general oversight role, including of annual appropriations, it has provided little scrutiny of strategic issues of performance, preferring instead to focus on administrative errors that might embarrass the government. But there have been some notable exceptions to the narrow focus and interests of parliament, and these have involved parliamentary committees inquiring into issues such as the Financial Management Improvement Program and service quality., These have been useful in emphasizing performance issues, such as the impact of government programs on their ultimate clients, to departments and to public servants more generally. Finally, there is a number of success factors that are often taken for granted in Australia, but that become evident when making cross-national comparisons, particularly with developing countries that do not possess a well-developed public service infrastructure: * strong institutional and human capacity in the public sector; * well-developed management capacity; * public service managers with a reputation for integrity, honesty and impartial advice; * well-developed accounting standards and systems; * a tradition of transparency and accountability in the conduct of government business; and * a credible and legitimate political executive. Parliament of Australia 1990. 193 ANNEX 14 Current Developments and Prospects The new government, elected in early 1996, has expressed considerable unhappiness with the efficiency of the federal public service and considers it to be rule-bound and caught up in red tape. It has also noted with dismay, evaluations that have shown public service efficiency to be lagging considerably behind that of private sector best practice, and these comparisons have strengthened its resolve to seek a smaller, more efficient public sector." The government has therefore embarked on a wave of major public sector management reforms. The new Cabinet has directed that Ministers and their departments review the nature and extent of continuing need for existing government programs; and it has expressed the expectation that competitive tendering and contracting processes be applied by departments to their programs wherever possible. These types of review require a close scrutiny of performance to be successful. In particular, the application of competitive tendering and contracting necessitates a clear understanding of program objectives, followed by ex ante assessments of the performance of alternative tenders (in-house and external). Once contracts have been let, there is a need for ongoing performance scrutiny, and at the completion of contracts, a review of past performance; this information then feeds back into decisions about the next round of contracts which are to be put out to tender. Evaluation of performance is central to all of these activities. The government's initiatives are part of a strong push towards commercialization and the private sector provision of public services, and have already resulted in a significant downsizing in the number of public servants, with further larger reductions in prospect. Under some authoritative scenarios, in a decade, the size of the public service could be only a fraction of its recent levels. The government is also taking steps to achieve sharper accountability for public service managers, together with fewer centralized controls and fewer formal requirements, partly via a strongly devolutionary approach. Departmental CEOs, who are on employment contracts, will be required to perform to high standards. These expectations are being made more explicit in legislation on financial management and accountability currently before parliament, and will increase pressure on CEOs to ensure high standards of corporate governance.5 This may create both the scope and a requirement for the national audit office to take a greater quality assurance role in performance measurement and reporting than in the past. Service delivery agencies will be required to set explicit customer service standards, with actual performance being reported publicly, including to parliament. The focus on performance will be further enhanced by the government's decision to adopt accrual accounting-this will facilitate scrutiny and comparisons of departmental performance. Changes to output/outcomes reporting are also in prospect, and these would seek to marry the output specification and focus of governments, such as those of New Zealand and several of the Australian states and territories, with the outcomes and performance focus of the federal Australian government. This development would also invite closer scrutiny of departments' planned and actual performance. " Reith 1996: MAB/MIAC 1996. The government has also favorably acknowledged the existence of "a system of performance management and program budgeting based upon an explicit evaluation of outcomes". (Reith 1996, p.ix) See ANAO 1997. 194 ANNEX 14 Collectively, the latest wave of reforms are likely to result in very considerable changes in performance management, accountability and reporting. For these reforms to be successful, however, there will need to occur a high level of scrutiny of departmental and CEO performance. This environment helps explain and put in context Cabinet's recent agreement to the replacement of the formal requirements under the evaluation strategy by a principles-based approach. This approach emphasizes the use of evaluations and other performance information for performance management purposes, including links with corporate and business planning and the other reform initiatives now underway. Thus the approach in one sense reaffirms the merits of the Financial Management Improvement Program and of program budgeting. The new approach continues to emphasize the advantages in planning, conducting, reporting and utilizing evaluation findings, the main difference with the previous evaluation strategy now being the absence of formal requirements." How should the new approach be viewed? If the judgment is made that at least part of the success of the requirement-based approach was that it mandated evaluation activity, then there will be some risk with a new approach that intentionally excludes such formal requirements. But a counter-argument is that the focus of concern should be with outcomes, not with processes to achieve them. If the public sector management framework is such that it achieves a strong focus on performance and outcomes, then this provides support to devolutionary approaches that provide management with the autonomy to achieve this performance in any manner it chooses. Put another way, in a government where performance and output measurement are strengthened, and there is greater accountability for results, there is scope to provide departments with greater autonomy and flexibility." One way of interpreting the new approach to evaluation is that it implicitly accepts performance measurement as an integral part of performance management"-reflecting a philosophy that if the environment of public sector governance is strongly conducive to evaluation being conducted and used, then that will happen. Thus the emerging public sector environment would be expected to be even more encouraging of evaluation than in previous years. The danger in this expectation is that it might mirror a similar but erroneous one in the early 1980s, when it was assumed that if the structural framework of public service management was 'correct', then an evaluative culture would almost automatically follow. The impact of the new reforms on the culture and management of the public service will partly depend on progress already achieved since the early 1980s. To the extent that an evaluation culture-including management commitment to review and learn from past performance, and an evaluation infrastructure-has already been achieved, it will enhance the speed and extent of impact of the new reforms. m And yet even some of the formal requirements will remain, but in a different guise: departments will continue to be issued guidelines for the reporting to parliament about their annual appropriations, and these will now include the need for summary statements of evaluation intentions (i.e., plans). In addition, guidelines for preparation of departments' annual reports will note the need to report on past performance, including as revealed by completed evaluations and other performance information. * This approach is advocated by the World Bank in its recent World Development Report (World Bank 1997). * In contrast, it could be argued that the earlier requirement-based approach had been added on to the then-existing suite of reform initiatives in the 1980s almost as an afterthought. If that is a fair interpretation, then it would mean that it had been a very effective afterthought. 195 ANNEX 14 Conclusions from the Australian Experience The requirements-based, formal evaluation strategy in Australia constituted a model of central force-feeding to help ensure the planning, conduct, quality and use of evaluation. It reflected the belief that managers would not do this if left to their own devices. Formal requirements and a whole-of-government approach helped to kick-start the process and achieve significant momentum, but it is worth bearing in mind that 'winning hearts and minds' cannot be mandated. It is not formal rules and requirements that determine the extent of the conduct and use of evaluation findings-it is the commitment of individuals and their institutions, and the nature of their understanding and motivation. The recent move to a principles-based approach reflects the evolution of governance arrangements and the particular circumstances of the Australian scene. This evolution represents a migration from tight to loose controls over departments and their conduct. The Australian experience provides a wealth of lessons, which can be drawn from. However, although it is possible to identify features of that system that have contributed to the substantial success achieved so far, this does not necessarily mean that these success factors are preconditions for success. Some of the key success factors have been: * macroeconomic pressures that have led to tight budgets and a priority on finding ways of achieving better value for money; * a whole-of-government approach to help achieve and maintain momentum in ECD; * a powerful department (DoF) willing to champion evaluation, react to changing circumstances and identify new opportunities for influence and development; * having a second central agency (the Auditor-General) willing to prompt and prod departments to focus on evaluation and performance management more broadly; * a budget agency (DoF) able to link evaluation into both the budget process and into program management reforms; * the implementation of a number of related public sector management reforms that emphasize bottom-line results and outcomes; * the support of Cabinet and a number of key Ministers, and the emphasis they have placed on having evaluation findings available to assist their decision-making; * the priority given to evaluation in several large and important line departments, which has helped to highlight and legitimize it; and * the devolutionary approach to evaluation within line departments, which has helped to mainstream evaluation as one of their core activities, together with internal quality assurance processes. 196 ANNEX 15 ANNEX 15: THE UNITED KINGDOM'S NEXT STEP AGENCIES The United Kingdom "Next Steps" program establishing Executive Agencies separate from government departments has attracted widespread international interest. As of November 1, 1995 there were 109 Executive Agencies and, additionally, two departments operating fully on Executive Agency lines (Table A15.1). These agencies employed 363,880 civil servant, or about two-thirds of the total. By 1997, the total employed in executive agencies is expected reach 75 percent of all civil servants. Each agency has a framework document, which sets out objectives and performance targets. Agencies are reviewed every five years to consider whether they still serve their purpose, whether they could be privatized as an enterprise, or whether they could be abolished. Table A15.1: United Kingdom: Executive agencies by size, 1995 Executive Agency Staff Social Security Benefits Agency 66,650 Inland Revenue 54,560 Employment Service 39,850 HM Prison Service 38,935 HM Customs and Excise 24,130 Defense Evaluation and Research Agency 11,680 Court Service 9,845 RAF Maintenance Group Defense Agency 8,945 Social Security Contributions Agency 8,900 HM Land Registry 8,510 RAF Training Group Defense Agency 7,410 Army Base Storage and Distribution Agency 6,400 Social Security Child Support Agency 5,985 Naval Recruiting and Training Agency 5,375 Sub-total 297,175 Other agencies 66,705 TOTAL 363,880 The chief executive is not a permanent civil servant, but on contract. The chief executive is responsible for the agencies' performance to the minister, but the minister remains responsible for policy. One of the major attractions of the Next Steps approach is a clearer focus on the performance of the agency, which, in turn, can be the basis for providing greater authority to the agency for management decisions. A precondition for efficient operation of the Next Steps agencies was the 1984 Financial Management Initiative which greatly improved accountability over financial resources, and the National Audit Act of 1984 which expanded audits to include efficiency and effectiveness audits. 197 BIBLIOGRAPHY ANAO (Australian National Audit Office). 1991. Implementation of Program Evaluation - Stage 1. Efficiency Audit Report No. 23, 1990-91. Canberra: Australian Government Publishing Service (AGPS). . 1991. Evaluation in Preparation of the Budget. Efficiency Audit Report No. 13, 1991-92. Canberra: AGPS. . 1992a. Program Evaluation in the Departments of Social Security and Primary Industries and Energy. Efficiency Audit Report No. 26, 1991-92. Canberra: AGPS. . 1992b. Auditing Program Evaluation - ANAO Performance Auditing Guide. Canberra: AGPS. . 1 992c. Department of the Treasury - Procedures for Managing the Economic Policy Program. Efficiency Audit Report No. 36, 1991-92. Canberra: AGPS. _ . 1993. Program Evaluation - Strategies, Practices and Impacts - Industry, Technology and Regional Development Portfolio. Efficiency Audit Report No. 35, 1992-93. Canberra: AGPS. . 1996. Performance Information - Department of Employment, Education, Training and Youth Affairs. Performance Audit Report No. 25, 1995-96. Canberra: AGPS. . 1997a. Applying Principles and Practice of Corporate Governance in Budget FundedAgencies. Canberra: AGPS. . 1997b. Program Evaluation in the Australian Public Service. Performance Audit Report No. XX, 1997-98. Canberra: AGPS. ANAO/DoF (Department of Finance). 1996. Performance Information Principles. Canberra: ANAO. Auditor General of Canada. 1993. Report of the Auditor General of Canada to the House of Commons. Chapter 8: Program Evaluation in the Federal Government; Chapter 9: Operation of Program Evaluation Units; Chapter 10: The Program Evaluation System - Making it Work. Ottawa. . 1996. Report of the Auditor General of Canada to the House of Commons - Chapter 3: Evaluation in the Federal Government. Ottawa. Barrett, Pat, Auditor-General of Australia. 1996a. "Some Thoughts About the Roles, Responsibilities and Future Scope of Auditors-General." Address at the Australian Society of Practising Accountants Annual Research Lecture, Canberra, November 14. . 1996b. "Performance Standards and Evaluation." Address at the Institute of Public Administration of Australia National Conference, Reshaping the Old: 198 Charting the New - Public Management in the 1990s, Melbourne, November 20- 22. Commonwealth of Australia. 1996. 1996-97 Budget Statement 3: Overview. Canberra: AGPS. DEET/DoF (Department of Employment, Education and Training/Department of Finance). 1994. Review of the Evaluation Function in DEET. DEET/DoF, Canberra. DEETYA (Department of Employment, Education, Training and Youth Affairs). 1996. Budget Initiatives. Canberra: DEETYA. Dixon, Geoff. 1993. "Managing Budget Outlays 1983-84 to 1992-93." In Brian Galligan, ed., Federalism and the Economy: International, National and State Issues. Canberra: Federalism Research Centre, Australian National University. DoF (Department of Finance). 1988. FMIP Report. Canberra: AGPS. . 1991. Handbook ofCost-Benefit Analysis. Canberra: DoF. . 1993. "The Cost of Evaluations: the Findings of a Pilot Study." DoF Working Paper, Canberra. . 1994a. Doing Evaluations - a Practical Guide. Canberra: DoF. . 1994b. The Use of Evaluation in the 1994-95 Budget. Finance Discussion Paper, DoF, Canberra. . 1995. "Reasons Why Evaluation Should Be Done and Why Finance Should Be Involved." DoF Working Paper, Canberra. . 1996. Performance Information Review: Overview of the First Year of the Review. Canberra: DoF. DoF and Public Service Board. 1986. Evaluating Government Programs - A Handbook Canberra: AGPS. Duckett, Mary. 1995. Performance Reporting in Commonwealth Annual Reports. Report for the Department of Finance. Canberra: DoF. Funnell, Sue. 1993. Effective Reporting in Program Performance Statements. Study for the Department of Finance. Canberra: DoF. Keating, Mike, and Malcolm Holmes. 1990. "Australia's Budgeting and Financial Management Reforms. " Governance: an International Journal of Policy and Administration 3(2): 168-185. Mackay, Keith. 1994. "The Australian Government's Evaluation Strategy: a Perspective From the Center." Canadian Journal of Program Evaluation 9(2): 15-30. MAB/MIAC (Management Advisory Board/Management Improvement Advisory Committee). 1995. Achieving Cost Effective Personnel Services. MAB/MIAC Report No. 18. Canberra: AGPS. 199 Parliament of Australia, House of Representatives Standing Committee on Finance and Public Administration. 1990. Not Dollars Alone - Review of the Financial Management Improvement Program. Parliament of Australia, Canberra. Parliament of Australia, Senate Finance and Public Administration References Committee. 1995. Service Delivery. Report from the Committee on Service Delivery by the Australian Public Service. Parliament of Australia, Canberra. Reith, Peter, Minister for Industrial Relations and Minister Assisting the Prime Minister for the Public Service. 1996. Towards a Best Practice Australian Public Service. Discussion Paper. Canberra: AGPS. Sedgwick, Steve. 1996. "Setting Ourselves for the Future." Address to DoF staff. Canberra: DoF. Senator Amanda Vanstone, Minister for Employment, Education, Training and Youth Affairs. 1996. Reforming Employment Assistance - helping Australians into real jobs. Canberra: AGPS. Task Force on Management Improvement. 1992. The Australian Public Service Reformed: an Evaluation of a Decade of Management Reform. MAB/MIAC, Canberra. Uhr, John and Keith Mackay, eds. 1996. Evaluating Policy Advice: Learning From Commonwealth Experience. Canberra: Federalism Research Centre (Australian National University) and DoF. World Bank. 1994. Evaluation Capacity Development - Report of the Task Force. World Bank, Washington, D.C. . 1997. The State in a Changing World: World Development Report 1997. New York, N.Y.: Oxford University Press. 200 References Fang Zheng, "An overall view of the key problems in current intergovernmental expenditure and revenue assignments", Difang Caizheng, 1997, No.8, pp. 6-11. Lou Jiwei and Ge Zhuying, "Experiments in Transfer Payments and Estimation of Gaps in Guizhou Province," paper presented at the International Conference on Intergovernmental Fiscal Relations in China, Guiyang, Guizhou, June 1997. Wong, Christine P.W., editor, Financing Local Government in the People's Republic of China, Hong Kong, Oxford University Press, 1997. Wong, Christine P.W. "Central-Local Relations in an Era of Fiscal Decline: The Paradox of Fiscal Decentralization in Post-Mao China," The China Quarterly (128), December 1991. Wong, Christine P.W., Christopher Heady and Wing T. Woo, Fiscal Management and Economic Reform in the People's Republic of China, Hong Kong, Oxford University Press, 1995. Xu Shanda, "Reform of China's Fiscal System: Present Situation and Trends", paper presented at the International Conference on Intergovernmental Fiscal Relations in China, Guiyang, Guizhou, June 1997.