102612 MINISTRY OF FINANCE PEOPLE’S REPUBLIC OF CHINA Subnational Debt Management and Restructuring LESSONS FROM INTERNATIONAL EXPERIENCE EDITORS Satu Kahkonen and Sudarshan Gooptu SUMMARIES AND PRESENTATIONS i The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations or to members of its Board of Executive Directors or the countries they represent. ii Table of Contents Acknowledgement ...................................................................................................................................... vii Foreword...................................................................................................................................................... ix Introductory Remarks at the Seminar ......................................................................................................... xi Introduction ................................................................................................................................................ xv Agenda .......................................................................................................................................................... 1 SECTION I - Getting the Context Right .......................................................................................................... 4 China’s Subnational Debts: Problems and Suggestions ............................................................................ 5 System Characteristics and Reform Logic of China’s Subnational Debts................................................ 15 SECTION II - Looking at the Past through Country Experiences .................................................................. 16 The United States: Regulatory Systems of Subnational Debt and Markets............................................ 17 Subnational Debt Management in Brazil and Mexico ............................................................................ 18 Subnational Debt: The Case of the Russian Federation .......................................................................... 25 Subnational Debt Restructuring in India ................................................................................................. 30 Fiscal Rules and Debt Restructuring: The EU Experience ....................................................................... 35 SECTION III - Addressing Subnational Indebtedness Problems – Distilling Lessons from Country Experiences ................................................................................................................................................. 42 Priorities in the Short and Medium-term for China’s Subnational Government Debt: Good Practices and Options ............................................................................................................................................. 43 Debt Restructuring, Institutions and Markets ........................................................................................ 57 Lessons for Borrowing Policy and Subnational Risks Management ....................................................... 58 Fiscal Adjustment and Subnational Borrowing Regulations ................................................................... 73 SECTION IV - Roundtable Discussion on Options and Subnational Reform Priorities for China - Learning from Cross Country Lessons........................................................................................................................ 91 Round-table Discussion: “Options and Subnational Reform Priorities for China - Learning from Cross Country Lessons” .................................................................................................................................... 92 SECTION V - Concluding Remarks ............................................................................................................... 95 Authors’ Biographies................................................................................................................................. 101 iii Section I: Getting the Context Right  “China’s Subnational Debts: Problems and Suggestions”. Dr. Liu Shangxi, Director of Research Institute for Fiscal Science, Ministry of Finance, China  “System Characteristics and Reform Logic of China’s Subnational Debts”. Dr. Qiao Baoyun, Dean of China Academy of Public Finance and Public Policy, Central University of Finance and Economics Section II: Looking at the Past through Country Experiences  North America (USA) – “The United States: Regulatory Systems of Subnational Debt and Markets”. Dr. Lili Liu, World Bank, Governance Global Practice, EFI Vice-Presidency  Emerging Markets: “Subnational Debt Management in Brazil and Mexico”. Dr. Fernando Blanco, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency  Emerging Markets: “Subnational Debt: The Case of the Russian Federation”. Dr. John Litwack, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency  Emerging Markets: “Subnational Debt Restructuring in India”. Ms. Abha Prasad, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency  European Union (EU): “Fiscal Rules and Debt Restructuring” (Greece and Ireland). Dr. Christoph T. F. Ungerer, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice- Presidency Section III: Addressing Subnational Indebtedness Problems – Distilling Lessons from Country Experiences  “Priorities in the Short and Medium-term for China’s Subnational Government Debt: Good Practices and Options”. Prof. Roy Bahl, Regents Professor of Economics and Dean, Emeritus, Andrew Young School of Policy Studies, Georgia State University, USA  “Debt Restructuring, Institutions, and Markets”. Dr. Lili Liu, World Bank, Governance Global Practice, EFI Vice-Presidency  “Lessons for Borrowing Policy and Subnational Risks Management”. Dr. Sudarshan Gooptu and Dr. Karlis Smits, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice- Presidency  “Fiscal Adjustment and Subnational Borrowing Regulations”. Prof. Jorge Martinez-Vazquez, Andrew Young School of Policy Studies, Georgia State University, USA Section IV: Roundtable Discussion on Options and Subnational Reform Priorities for China - Learning from Cross Country Lessons  Dr. Qiao Baoyun, Dean of China Academy of Public Finance and Public Policy, Central University of Finance and Economics, USA  Prof. Roy Bahl, Andrew Young School of Policy Studies, Georgia State University, USA  Prof. Jorge Martinez-Vazquez, Andrew Young School of Policy Studies, Georgia State University, USA iv  Dr. Satu Kahkonen, Director for Asia and Europe, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency Section V: Concluding Remarks  Mr. Chen Xinhua, Deputy Director-General, Budget Department, Ministry of Finance, China  Dr. Satu Kahkonen, Director for Asia and Europe, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency, World Bank  Mr. Xu Xiang, Deputy Director-General of Human Resources Development Centers, Ministry of Finance, China v vi Acknowledgement First and foremost, we would like to appreciate the long-standing cooperation between the Ministry of Finance and the World Bank Group in China in thinking through and working together to tackle the most complex and pertinent economic policy issues facing the country at various junctures of its highly successful development experience. We are honored to have this opportunity to jointly co-host this Subnational Debt and Restructuring Summit with the Ministry in Nanning, and for the rapid production of this e-book to synthesize its proceedings and make it widely available to all. We look forward to many such joint knowledge sharing activities and joint research activities in future. We would like to thank our hosts in Nanning, the Department of Finance of Guangxi Zhuang Autonomous Region, especially DG Huang Weijing and his staff who warmly welcomed us there, and to Messrs. Wu Zhenpeng, Xu Xiang, Chen Xinhua, Mao Wenyong and Gu Ming from the Ministry of Finance and their staff in Beijing, who worked very effectively together with us to make this Summit a success. The event and e-book would not have been possible without the instruction of H.E. Mr. Xu Hongcai, Assistant Minister of Finance, and the excellent cooperation from the Department of Budget, Department of International Cooperation, Department of International Economic and Financial Cooperation, Human Resources Development Center of Ministry of Finance, as well as the contributing expert panelists, thought leaders on the subject from China and abroad, and from officials from the Government and the World Bank experts who helped make the discussions at the summit highly relevant and substantive to the issues at hand. Contributions by our colleagues Chorching Goh, John Litwack, Karlis Smits, Lili Liu and Christoph Ungerer in designing the event agenda and helping with its delivery along with Mmes. Yu Jinhong, who provided translation services, and Wang Lin, who helped with coordination, are greatly appreciated. Mmes. Yang Lin and Ivana Ticha worked together as a seamless team with all the administrative support during the entire process to make both the Nanning Summit and e-book a reality from the World Bank’s side. Satu Kahkonen Sudarshan Gooptu vii viii Foreword Jan Walliser Vice-President, Equitable Growth, Finance and Institutions (EFI) Practice Group Vice-Presidency, World Bank The Global Financial Crisis made clear that debt can spiral out of control in both developed and developing countries. An important lesson from this and previous crises has been that it is much less costly to manage debt levels before the onset of a crisis than after the crisis has already hit. In a country as large as China, the debt of cities, provinces, and other subnational governments is important. It has been important for growth: subnational loans have financed roads, bridges, utilities infrastructure and other public investment that has driven GDP growth and drastically reduced poverty in China in the last decade. But as the economy is shifting to new growth drivers, managing debt that previously catalyzed the economy is the next, crucial step. It is vital to the country’s – and, some might argue, the world’s – economic stability. The Government of China is now looking into ways local government can streamline and make more effective the management of public finances and subnational debt. One significant step toward a modern fiscal system is the Budget Law, signed in August 2014. This law lays the foundation for the overhaul of the fiscal system. It aims to improve the quality of budget management, encourage local governme nts’ debt sustainability, and facilitate the channeling of debt in local government financing vehicles to more transparent bond markets. These measures will also support efforts by the central government to align local governments' objectives more directly with an agenda for effective and efficient public service provision in China. Subnational debt problems are not unique to China or to developing countries. They have recently been observed in the U.S. – in Puerto Rico, a U.S. territory, and Detroit, a major U.S. city – and in the United Kingdom (City of London, Scotland), among others. This e-book provides examples to help policymakers in heading off those situations. It chronicles international experiences and lessons learned on how countries have successfully managed their subnational debt. It looks at how they undertook effective fiscal adjustment measures - not only to restructure the troubled subnational debts but also to adjust their subnational economies through better fiscal and debt management - with varying levels of involvement from the central government. Notable examples of successful measures include policies that reformed tax collection and administration, as well as monitoring and managing fiscal risks from contingent liabilities, which can be particularly important in a country such as China, where state-owned enterprises make such risks more complex. The e-book also addresses changing intergovernmental fiscal relationships, and tactics that involve imposing fiscal rules in order to maintain fiscal and debt sustainability over time. The materials in this e-book were first presented at the “Subnational Debt Summit” held in Nanning, China on October 22, 2015. I am confident that readers will benefit from the rich, international ix experiences presented in this work. I hope these lessons help inform China’s practices and thinking towards its own reforms going forward. x Introductory Remarks at the Seminar Wu Zhenpeng Director-General, Human Resources Development Center, Ministry of Finance, People’s Republic of China Learn from International Experience, Enhance Local Government Debt Management (October 22, 2015) Distinguished experts domestic and from overseas, ladies and gentlemen, Good Morning! Currently, local government debt issue is a generalized economic challenge among countries of the world. Governments are facing the threat of increasing local debt risks. Today, the Department of Budget, Department of International Cooperation, Department of International Economic and Financial Cooperation, Human Resources Development Center of Ministry of Finance along with the World Bank are jointly holding this Local Debt Management International Seminar to discuss how to better exert the fiscal policy role and enhance local government debt management. I think that it’s very meaningful. In accordance with the seminar schedule and instructions from Mr. Xu Hongcai, Assistant Minister of Finance, also delegated by the Department of Budget, I will provide a brief introduction to China’s local government debt situation for your reference. I. Basic condition of local government debt management With over 30 years of reform and opening-up, our local governments through financing platform company and other debt finance methods have speeded up the urban and rural infrastructure construction and urbanization process, and spurred the development of public undertakings in education, culture, and health. However, local government debt over the years has been beset by problems mainly manifested in the lack of management standard, high financing cost, irrational maturity structure and risk potential in some regions and industries. In recent years, Chinese Government, with highly responsible attitude, has always taken measures to enhance local government debt management arrangement. On one hand, by cleaning up local government financing platform companies to regulate its borrowing practice; on the other hand, by issuing local government bond while seeking to establish standardized local government financing mechanism with some effect. At the same time, Ministry of Finance along with the National Development and Reform Commission, the People's Bank of China, the National Audit Office and the China Banking Regulatory Commission has earnestly implemented the arrangement of the State Council, actively researched and issued policies and measures on enhancing local government debt arrangement, and regulated local government financing platform company and local government debt management, obtaining considerable progress. The moves taken have played a positive role in successfully coping with the international financial crisis, managing inflation, implementing structural adjustment and advancing sustainable and healthy economic and social development. xi After filtering out and screening by the end of 2014, the total balance of local government debt stood at 15.4 trillion yuan; it’s estimated that by the end of 2015 national local government debt ratio will reach 86%, which matches with what Premier Li Keqiang pointed out in his speech at the 9th Summer Davos Forum held in September 2015 that Chinese governmental debt “risk is controllable.” II. Overall systems arrangement for local government debt management As the local government debt impacts the overall situation, the Chinese Government has attached great importance to it. In June 2014, the Central Government deliberated and approved the “Deepening Reform of Fiscal and Taxation System Overall Scheme,” setting forth the general requirements for regulating local government debt management. The amended Budget Law deliberated and approved by the Standing Committee of the National People's Congress (NPC) in August 2014 gave green light to local governments to borrow at a proper level, while making a systematic arrangement for local government debt management from the aspects of debt entity, borrowing methods, scope control, budget management, use of borrowed money, risk control, and accountability. In September 2014, the State Council published an important document titled “Opinions on Strengthening Management of Local Government Debt,” further clarifying the overall system arrangement for local government debt management. Grant permission to local government to raise fund in accordance with law. Provincial- level government can borrow within the scale approved by the State Council. Prefecture-level and county-level governments can commission the provincial- level governments to borrow, if necessary, to regulate local governments’ debt financing behavior. Local governments can issue municipal bonds to remove government financing function from financing platform company; at the same time, they can actively promote the cooperation mode between the government and social capital (via public-private partnership (PPP)) to attract social capitals participating in investment and operation of public welfare undertaking. To regulate local government debt borrowing procedures and use of fund, local government debt borrowing must be approved by the local Standing Committee of the People's Congress. Fund raised must be used for public welfare expenditure. To prevent and defuse local government debt risk, local government debt must be categorized and included in the management of local government budget. Set up risk assessment and early warning mechanism, emergency response mechanism and accountability system. The Opinion clearly marks the beginning of a major new initiative to cap and control the local debt problem by establishing an unified “borrowing, use and repayment” local government debt management mechanism, effectively playing a positive role in local government borrowing, and having an impact on guarding against budgetary financing risk and promoting sustainable and healthy development of economy. III. Key measures of regulating local government debt management Since this year, the Ministry of Finance has taken the following steps to regulate local government debt management. Along with relevant departments, it conducted an inventory and screening on total local government debt by the end of 2014 and reported to the State Council and NPC Standing Committee. In August 2015, The Standing Committee of China’s National People’s Congress approved a 16 - trillion- yuan debt for the local governments; and Ministry of Finance timely allocated the approved regional debt quota. The amount made available via a special bond program has been increased to 600 billion yuan, which was mainly to finance government-subsidized housing like shanty town revamping, ordinary xii highway projects, and other key areas designated by the State Council. It expanded a local-debt refinancing quota to 3.2 trillion yuan to lower interest burden. Under this program, China’s local governments can swap new bonds for high-interest local debt, giving them more breathing room to meet their obligations and concentrating capital on key project construction. At present, China is at the phase of “rapid urbanization” while the conflict of capital demands for local infrastructure building and local government budget constraint gets more acute; short-term response and long-term system building must be implemented simultaneously, combining release and block, so as to establish long-term mechanism of regulated local government debt financing. Next step, the Ministry of Finance will research and enact related systems and methods of risk assessment & early warning, total debt incorporated into budget management or other debt management to insure the implementation of policies. Now the Party Central Committee, State Council and Party Group of the Ministry of Finance have put forth higher demands for local government debt management. Finance and budgetary colleagues have a long way to go. Therefore, I hope that finance department at various levels should enhance awareness, shoulder responsibility and press ahead with the implementation of local government debt management policy in a steady and incremental fashion. Under the backdrop of accelerating globalization, interactions among countries has become ever more frequent; as a result, it is critically important to strengthen exchange and collaboration on the knowledge and academic field. The Ministry of Finance has a great tradition, that is, they often listen to experts’ opinion to enhance the scientific and effective nature of financial policy. The main purpose of the seminar jointly held by the Ministry of Finance and the World Bank is to listen to your insightful opinions. I hope that experts and attendees presented here will take this opportunity to conduct in-depth discussion on key theoretic and real-life issues regarding governmental debt, freely exchange ideas, make suggestions and offer your knowledge and wisdom to advance scientific arrangement of local government debt. In recent years, Human Resources Development Center in synch with the new round of finance and taxation system reform has been dedicated to co-develop international knowledge communication platforms with international organizations like the World Bank to serve the core tasks of finance and foster the development of fiscal cadres. My colleagues at the Human Resources Development Center all hope to make their own contribution to the frontline area of local government debt management. Through international cooperation, forge “knowledge base” for China’s finance and taxation system reform, it will become a “think tank” of top-level policy design and “refueling station” for financial workers to borrow from global experience and enhance their own overall quality, spur the implementation of key reform at budget department and related departments, and push forward development of the new round of finance and taxation system reform. I remember Wei Zheng, the famous Tang Dynasty court official and politician once said: “The difficult thing is not just to know something, but to do it; the difficult thing is not even to do something, but to be persistent in doing it”. This reminds us of not only dedication to learning but also application theory to practice, bold in action; instead of doing things by halves, we should persist to the end as “greatness shines from start to finish”. The more precious quality of a scholar is his practice instead of learning. Thus, I hope that all attendees will value this opportunity to draw insights from international xiii experience, and gain new ways of thinking. More importantly, we can apply knowledge learned to future work, realizing “gain from learning and use from learning.” At last, on behalf of organizer of the seminar, I convey my sincere thanks to the World Bank which has lent time-tested support for China’s reform process, to experts domestic and overseas in attendance, and to the Department of Finance of Guangxi Zhuang Autonomous Region that has made meticulous preparations for this seminar. I wish a great success to “local debt management international seminar.” xiv Introduction Bert Hofman Country Director for China, Korea and Mongolia, World Bank In the aftermath of the global financial crisis, policymakers around the world are focusing once again on government debt sustainability. In China, subnational government debt is an important part of total government debt, and therefore deserves the attention that policymakers have paid to the topic. Subnational debt has played an important role in financing China’s impressive infrastructure that is the envy of the world. And subnational debt was instrumental in the economic stimulus that China so effectively staged after the global financial crisis, through which China maintained high levels of economic activity. Now that China’s economy has entered a “New Normal” a reassessment of the levels of Subnational Debt, the mechanisms by which such debt is accumulated and managed, and the sustainability of subnational debt are being debated in China. It is with this in mind that we were honored to co-host this joint International Workshop with the Ministry of Finance and the Department of Finance of Guangxi Zhuang Autonomous Region on Subnational Debt Management and Restructuring in the beautiful City of Nanning in the Guangxi Zhuang Autonomous Region of China. Since the onset of Opening Up and Reforms in 1978, fiscal reforms have been an integral part of the government’s overall reform program. The “Eating from Separate Kitchens” reform of 1980 unleashed the initiative and enthusiasm of local governments, which drove many of the reforms that resulted in China’s high growth and poverty alleviation. The Tax Sharing Systems reforms of 1994 allowed central government to regain macroeconomic control, which was imperative in light of the runaway inflation of those days. The Budget Law of 1995 defined the role of the budget in a socialist market economy, while the inclusion of extrabudgetary funds in the budget at the end of the 1990s allowed governments at all levels to manage their resources in a more comprehensive manner. It is no surprise that the reforms of the 3rd plenary session of 18th Central Committee in 2013, which aim to “let the market play a decisive role in resource allocation” has major implications for fiscal policy as well. The amendments of the Budget Law approved last year can be seen in the spirit of that reform in China, and a crucial element of the new budget law is the way subnational government debt is defined and managed. Undoubtedly, the new Budget Law requires local governments to adjust, to regularize their use of debt finance, and more clearly distinguish between government and non-government debt. Although China’s overall subnational debt levels are manageable, for some local governments, the new budget law and the “New Normal” will also mean adjustment so that their debt remains sustainable and access to the capital markets to finance future growth is maintained. There is rich international experience on how countries have managed their subnational debt and how subnational fiscal adjustment has been used to maintain sustainable debt levels. The conference reviewed some of that experience, ranging from States in Brazil and Mexico and India to adjustment in countries within the Eurozone. Beyond debt management and debt restructuring, the conference discussed what changes in tax structure, revenue mobilization, expenditure assignment and revenue xv sharing could be used to ensure that subnational debt remains sustainable and subnational governments can play their role essential in China’s national development. This e-book starts with a summary and presentation of the Chinese perspective on this subject of Subnational Debt in their own words (Session 1). In doing so, this session at the Nanning workshop provided a good opportunity for foreign experts present there to learn, and the over 200 Chinese officials from subnational levels of government to exchange views and share their own perspectives during and after the sessions. Session 2 introduced some international cases which are highly relevant to the current situation in China. Session 3 brought these international case studies and experts’ perspectives together to arrive at syntheses that narrowed down key issues which China needs to consider going forward in subnational debt management and restructuring. The roundtable discussion among international and Chinese experts on the way forward for China in Session 4 provided an illuminating discussion of this issue. We hope that this will inform specific policies that will be formulated to help local governments in China manage their subnational debts. In this way, fund raising by local governments can be ensured and will continue to play a role in China’s development. The international experiences and policy recommendations discussed in the workshop are relevant for many countries around the world, and I hope that with the publication of this e-book we will be able to support those countries in making their own policy choices in the context of their specific policy challenges and circumstances. xvi Agenda World Bank-Ministry of Finance, China Joint Workshop on SUBNATIONAL DEBT MANAGEMENT and RESTRUCTURING October 22, 2015. Nanning, Guangxi Zhuang Autonomous Region, P.R. CHINA 9:00 a.m. - 9:30 a.m. Opening Remarks Moderator: Mr. Chen Xinhua, Deputy Director-General, Budget Department, Ministry of Finance  Mr. Wu Zhenpeng, Director-General, Human Resources Development Center, Ministry of Finance, China  Mr. Huang Weijing, Director-General, Finance Department of Guangxi Zhuang Autonomous Region  Dr. Bert Hofman, Country Director for China, Korea and Mongolia, World Bank 9:30 a.m. - 10:00 a.m. Session 1: Getting the Context Right Introductory presentation by Chinese Govt./Researchers on “Current Subnational Debt Situation and related-Emerging Policy Issues” – (i) Central Government Perspective and (ii) Provincial/local Government Perspective Moderator: Mr. Xu Xiang, Deputy Director-General of Human Resources Development Center, Ministry of Finance, China  “China’s Subnational Debts: Problems and Suggestions”. Dr. Liu Shangxi, Director of Research Institute for Fiscal Science, Ministry of Finance  “System Characteristics and Reform Logic of China’s Subnational Debts”. Prof. Qiao Baoyun, Dean of China Academy of Public Finance and Public Policy, Central University of Finance and Economics, USA 10:00 a.m. - 12:45 p.m. Session 2: Looking at the Past through Country Cases This session will review good practices on Subnational Debt Management: Examples from Selected Advanced and Emerging Economies. (With special emphasis on Subnational Debt Restructuring and associated reforms undertaken in selected countries.) Moderator: Dr. Mathew Verghis, Practice Manager for East Asia and the Pacific, Macroeconomics and Fiscal Management, Equitable Growth, Finance and Institutions (EFI) Vice-Presidency, World Bank  North America (USA): “The United States: Regulatory Systems of Subnational Debt and Markets”. Dr. Lili Liu, World Bank, Governance Global Practice, EFI Vice-Presidency  Emerging Markets: “Subnational Debt Management in Brazil and Mexico”. Dr. Fernando Blanco, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency 1 10:50 a.m. - 11:05 a.m. Break  Emerging Markets: “Subnational Debt: The Case of the Russian Federation”. Dr. John Litwack, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency  Emerging Markets: “Subnational Debt Restructuring in India”. Ms. Abha Prasad, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency  European Union (EU): “Fiscal Rules and Debt Restructuring” (Greece and Ireland). Dr. Christoph T. F. Ungerer, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice- Presidency 12:45 p.m. - 2:00 p.m. Lunch Break 2:00 p.m. - 3:45 p.m. Session 3: Addressing Subnational Indebtedness Problems – Distilling Lessons from Country Experiences This session will provide a synthesis of key lessons that emerge from the selected country experiences on subnational debt-related regulations, institutions, borrowing framework, capital market linkages, insolvency regimes, fiscal rules and fiscal risk management.  “Debt Restructuring, Institutions, and Markets”. Dr. Lili Liu, World Bank, Governance Global Practice, EFI Vice-Presidency  “Lessons for Borrowing Policy and Subnational Risks Management”. Dr. Sudarshan Gooptu, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency  “Fiscal Adjustment and Subnational Borrowing Regulations”. Prof. Jorge Martinez-Vazquez, Andrew Young School of Policy Studies, Georgia State University, USA 3:45 p.m. - 4:00 p.m. Break 4:00 p.m. - 5:45 p.m. Session 4: Roundtable Discussion on Options and Subnational Reform Priorities for China - Learning from Cross Country Lessons Moderator: Dr. Bert Hofman, Country Director for China, Korea and Mongolia, World Bank  Mr. Qiao Baoyun, Dean of China Academy of Public Finance and Public Policy, Central University of Finance and Economics  Prof. Roy Bahl, Andrew Young School of Policy Studies, Georgia State University, USA  Prof. Jorge Martinez-Vazquez, Andrew Young School of Policy Studies, Georgia State University, USA  Dr. Satu Kahkonen, Director for Asia and Europe, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency, World Bank 5:45 p.m. - 6:00 p.m. Concluding Remarks  Mr. Chen Xinhua, Deputy Director-General, Budget Department, Ministry of Finance, China  Dr. Satu Kahkonen, Director for Asia and Europe, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency, World Bank 2  Mr. Xu Xiang, Deputy Director-General of Human Resources Development Center, Ministry of Finance, China 3 SECTION I - Getting the Context Right “China’s Subnational Debts: Problems and Suggestions”. Dr. Liu Shangxi, Director of Research Institute for Fiscal Science, Ministry of Finance “System Characteristics and Reform Logic of China’s Subnational Debts”. Prof. Qiao Baoyun, Dean of China Academy of Public Finance and Public Policy, Central University of Finance and Economics, USA 4 China’s Subnational Debts: Problems and Suggestions Liu Shangxi1 I. Introduction Overall Situation The audit results given by China’s National Audit Office at the end of June 2013 show that subnational debts comprise of three categories: The debts to be repaid by local governments, the debts guaranteed by local governments, and the debts that may be relieved by local governments. As shown in Figure 1, the debt levels verified at the end of June 2013 under these three categories were 10.8 trillion yuan, 2.6 trillion yuan and 4.3 trillion yuan, respectively. The first category are direct debts of governments, while the second and third categories are the contingent debts of governments. The direct debts are of a larger scale than the contingent debts. Among contingent debts, the scale of contingent debts is smaller than that of (moral) contingent debts. Figure 1. The overall scales of three categories of subnational debts (Unit: 100 million yuan) 120000 108859.17 100000 96281.67 80000 60000 43393.72 37705.16 40000 24871.29 26655.77 20000 0 Debts to be Repaid by Governments Debts Guaranteed by Contingent政府负有担保责任 Other Relevant Debts May be 政府负有偿还责任 政府承担一定救助责任 Governments Relieved by Governments 2012年底 End of 2012 2013年6月底 End of June 2013 Situations at different government levels Finding of the debt audit results as of the end of June 2013 show that the debts to be repaid by local governments are mainly at municipal, district and county levels. Municipal level accounts for 44.5%, county and township levels account for 39.2%, while provincial level only accounts for 16.3%. Table 1. shows the main situation of debts to be repaid by local governments. 1 Research Institute for Fiscal Science, Ministry of Finance. 5 Table 1. Debts to be re-paid by local governments as of June 2013 (Unit: 100 million yuan) Balance of Debt Structure of Debt Percentage (100 million yuan) Debts to be Repaid by Governments 108859.17 60.85% Structure of Contingent Debts Guaranteed by Governments 26655.77 14.90% Government Other Relevant Debts That May be Relieved by Responsibilities Governments 43393.72 24.25% Provincial-level government debts 17780.84 16.33% Municipal-level government debts 48434.61 44.49% Tiered Structure County-level government debts 39573.6 36.35% Town/township-level government debts 3070.12 2.82% Financing Platform Companies 40755.54 37.44% Departments and Institutions of Local Government 30913.38 28.40% Institutions with Outlay Subsidies 17761.87 16.32% Main Structure Wholly State-owned or State Holding Enterprises 11562.54 10.62% Self-sufficient Institutions 3462.91 3.18% Public Institutions 1240.29 1.14% Other Units 3162.64 2.91% Bank Loans 55252.45 50.76% BT 12146.3 11.16% Trust Financing 7620.33 7.00% Structure of Fund Issuing Bonds 11658.67 10.71% Sources Loans from Other Units and Individuals 6679.41 6.14% Payable Payments 7781.9 7.15% Others 7720.11 7.09% Payable Money 7670.4 7.05% Municipal Construction 37935.06 34.85% Transportation Infrastructure Construction 13943.06 12.81% Land Expropriation and Reserve 16892.67 15.52% Science and Technology, Culture and Health 4878.77 4.48% Structure of Fund Construction for Agriculture, Forestry, Water 3.75% Usage Conservancy 4085.97 Ecological Construction and Environmental 2.96% Protection 3218.89 Low-income Housing 6851.71 6.29% Industry and Energy 1227.07 1.13% Others 12155.57 11.17% Repaid between July and December of 2013 24949.06 22.92% Repaid in 2014 23826.39 21.89% To be Repaid in 2015 18577.91 17.07% Term Structure To be Repaid in 2016 12608.53 11.58% To be Repaid in 2017 8477.55 7.79% To be Repaid in 2018 and thereafter 20419.73 18.76% Total 108859.17 100% Source: Compiled by Ministry of Finance in accordance with the Audit Results of Nationwide Government Debts (2013 No. 32). 6 Figure 2. Composition of provincial-level, municipal-level and county-level subnational debts as of the end of June 2013 (Unit: 100 million yuan) 60000 50000 40000 30000 20000 10000 0 Repayment Guarantee Relief Repayment Guarantee Relief Repayment Guarantee Relief 偿还责任 Responsibility Responsibility 担保责任 Responsibility 救助责任 偿还责任 Responsibility 担保责任 救助责任 Responsibility 偿还责任 Responsibility 担保责任 Responsibility 救助责任 Responsibility Responsibility 省级 Provincial-level 市级 Municipal-level 县乡级 County- and Township-level Source: Compiled by Ministry of Finance in accordance with the Audit Results of Nationwide Government Debts (2013 No. 32). Situation of borrowers As for borrowers, apart from debts guaranteed by local governments, the debts incurred through financing platform companies and shall be repaid or relieved by governments are of the largest scale. The departments and institutions of local governments are the borrowers with the largest scale of guaranteed debts. Figure 3 shows the three categories of debts by borrowers. It shows that subnational debts that are financing platform companies as well as local government departments and institutions account for approximately 66% of the total. 7 Figure 3. Composition of main borrowers of subnational debts as of the end of June 2013 (Unit: 100 million yuan) 45000 40000 35000 30000 25000 20000 15000 10000 Repayment Repayment Repayment Responsibility 5000 Responsibility Responsibility 0Repayment Guarantee Relief Guarantee Relief Guarantee Relief Guarantee Relief Responsibility 偿还责任 Responsibility 担保责任 救助责任 偿还责任 Responsibility 担保责任 救助责任 Responsibility 偿还责任 Responsibility 担保责任 Responsibility Responsibility 救助责任 偿还责任 担保责任Responsibility Responsibility 救助责任 Departments and Institutions of Institutions with Outlay Subsidies Public Institutions 融资平台公司 Financing Platform 地方政府部门和机构 Local Governments 经费补助事业单位 公用事业单位 Companies Source: Compiled by Ministry of Finance in accordance with the Audit Results of Nationwide Government Debts (2013 No. 32). Note: The debts of local government departments and institutions fall within the scope of debts to be repaid and guaranteed by governments, but not to be relieved by governments. Debt-repayment situation According to the audit results of 2013, the subnational debts to be repaid in 2015 and thereafter are shown in the following table. Overall, the three categories of debts that will mature in 2015 are of the largest scale in the coming three years. Table 2. Subnational debts that will mature in 2015 and thereafter (Unit: 100 million yuan) 2015 2016 2017 2018 and Thereafter Amount Percentage Amount Percentage Amount Percentage Amount Percentage To be Repaid by 18577.91 17.06% 12608.53 11.58% 8477.55 7.79% 20419.73 18.76% Governments Guaranteed by 3198.42 12.00% 2606.26 9.78% 2298.6 8.62% 11706.75 43.92% Governments May be Relieved 5994.78 13.81% 4206.51 9.69% 3519.02 8.11% 16669.05 38.41% by Governments Source: Compiled by Ministry of Finance in accordance with the Audit Results of Nationwide Government Debts (2013 No. 32). 8 The consequences of these subnational debts are opaque risks and associated uncertainties arising from them. As a result, markets, public opinion, and even the international community are making various conjectures and predictions on China’s local government debt risks. This has enhanced the uncertainty and public risk perceptions on the economy, society and even China’s development prospects. If debt is likened to dynamite, the most terrible thing is not the dynamite itself, but the unclear explosive conditions of the dynamite, or that the risk is obscured. Under such circumstances, the risk may be either exaggerated, which will result in panic; or be underestimated, which could result in a crisis of unexpected proportions. Both exaggeration and underestimation of risk means that the risk is out of control, the result of which could be the accumulation and spread of risk and unrecoverable losses. The opaque risks of subnational debts mainly arise from the following:  First, there are no regular statistics and release of subnational debt data.  Second, there are no governmental accounting standards for the confirmation and measurement of subnational debts.  Third, the debt risk management system is incomplete; there lacks an effective management system for identifying, analyzing, preparing for and responding to risks. In particular, there lack effective early warning and response tools for the management of liquidity risks.  Fourth, the debt management system is incomplete; the authority over, control of and responsibility for risks are not very clear. The authorities and responsibilities between the central government and local governments, among local governments at various levels, between government and financing platform, and among government-controlled SOEs need to be clarified. The system of “risk egalitarianism” remains to exist.  Fifth, there is no connection between local debt management and budget management. It is up to the central government to decide how much local governments can borrow by issuing bonds. Generally, the amounts can only be known after the NPC meetings. From the distribution of bonds issuance data to the completion of debt issuance, the earliest for the process to end will be May. It is not realistic to put existing subnational debts into budget management. The objects of budget management are revenue and expenditure flows, while the objects of debt management include not only flows but also stock, or the balance of debt. What can be put into budget management are the incremental part of debt balance and the interests paid during the year. And the contingent debts cannot be put into budget management. II. Relationship between the Reform of Fiscal and Taxation Systems and the Improvement of Local Debt Financing System Fiscal and taxation systems are a comprehensive concept, which include the fiscal relationship between the central government and local governments (customarily called the “fiscal system”), budget system and taxation system. Among them, the fiscal system is directly related to the financing of local debts. Theoretically, a country’s fiscal system includes three key factors: administrative powers, financial powers and financial resources. And the three corresponding aspects are: division between the administrative powers of the central government and local governments to form their respective 9 expenditure responsibilities; division between the financial powers of the central government and local governments, including taxation power, fee-collecting power and property power; and transfer payments between governments. The financing system of subnational debts is based on the national fiscal system and focused on the division of administrative powers between the central government and local governments. The specific contents of the financing system of local debts are the arrangements for the borrowing, usage and repayment of money among local governments at various levels. Looking from the tiered governance of a country, China has two governance tiers: national (central) governance and local governance. Correspondingly, there are also two levels in the fiscal system: One is the country’s fiscal system, defining the fiscal relationship between the central government and local governments; the other is the local fiscal system, defining the fiscal relationship among local governments at various levels. It is not difficult to see that, the local financing system is a part of the local fiscal system, involving the division of administrative powers among provincial-level, municipal-level, county-level and township-level governments. The financing system of local debts needs to be improved from two levels. One is to address what the central government should do and should not do, so as to clarify what local governments should do and should not do. These are the premise. The division of revenue on the basis of dividing administrative powers between the central government and local governments, while the division of expenditure responsibilities and the design of transfer payment system determine the financing needs of local debts. The other challenge is to address what the provincial-level, municipal-level, county-level and township- level governments should do and should not do. Local government is not a general concept. Contrary to debts of the central government, those of the local governments are distributed at various levels, and so are their debt-repaying responsibilities. In reality, the problems at the above-mentioned two levels have not been well addressed. Under the overall governance mode where “the central government makes decisions and local governments implement them”, local governments cannot decide what they should do or should not do, and must follow the instructions from the central government. If local governments can decide what they should do, they will generally do things that are favorable to localities. When coping with the international financial crisis in 2008, the central government requested local governments to take initiatives together, changing the past practice of the central government taking full responsibility in the 1998 financial crisis. As a result, local financing platforms and large amounts of debts were created. In China, it is local governments that drive the urbanization process; and urbanization-related infrastructure construction is mainly financed by borrowing. It is local governments that make stable growth possible. Under the circumstance of local fiscal expenditure accounting for 85% of nationwide fiscal expenditure, local governments are the main player for implementing policies for expanding investment expenditure or public consumption expenditure. The more central macro regulation relies on local governments, the more local governments will rely on borrowing. Looking from local perspective, there shows a pressure transmission system among the relationship of provincial-level, municipal-level, county-level and township-level governments. The pressure imposed by the central government will be transmitted downward level by level. Since provincial-level governments have more power, they can delegate many things to lower-level governments. Of local 10 debts, therefore, provincial-level governments account for less proportion than municipal-level and county-level governments do. But the debt-repaying responsibilities are not symmetric. When the municipal-level and county-level governments are unable to repay their debts, provincial governments have the responsibility to repay the said debts. After all, they cannot let grassroots fiscal systems go bankrupt. When the rural cooperative foundations went bankrupt in 1999, the provincial-level governments took the responsibility of repaying their debts. III. Local Government Debt Risk Assessment China’s subnational debts are accumulated over many years. Currently, the local government debt risk is generally controllable. However, some subnational debts pose certain risks. To comprehensively understand debt situation and risk, the National Audit Office of the People’s Republic of China (NAO) conducts three special audits of subnational debts. According to the audit report issued on December 30, 2013, by the end of 2012, China’s total local liabilities amounted to 39.43% of GDP, lower than 60% of public debt-to-GDP ratio, which is often noted as an international limit for debt control. Total debt was 113.41% of government fiscal revenue within the controlled debt ratio proposed by the International Monetary Fund. It could be concluded that the risk of China’s government debt is controllable generally, while the risk of government debts mainly lies in the local governments. Among subnational debts, municipal and county governments account for around 80%, financing platform companies and local government departments and institutions make up about 66%, bank loans and debts in disguised form account for about 82% and municipal construction, transportation, land resource collection and reserve, science, education, culture and public health, and low-income housing account for around 70%. The risks are larger in these areas. In terms of the growth of subnational debts, after June 2013, the growth of local government debt slows down. In the first half of 2014, NAO surveyed the debt of 9 provinces and 9 cities between the end of 2013 and the end of March 2014. The result showed that the growth rate of local government debt slowed to 3.79%, down seven percentage points than the average growth rate in the first six months of 2013. The debt growth rate slows down2. Generally, the risk of local government debt is under control. However, the risk is expanded currently due to increasing debt repayment pressure, and will in the future as a result of the rapid debt growth in the past and poor debt management. Also, the risk of local government is further expanded with current economic slowdown, declining the growth rate of financial income (including land income) and rigid growth of expenditure. IV. Countermeasures and Suggestions Basic Ideas After the revised Budget Law was deliberated and adopted in 2014, the establishment of standard local government debt financing system is orderly promoted in the whole country according to laws. The 2 NAO: The Growth of Local Government Debts Slows with Risks under the Control: http://money.163.com/14/0824/17/A4E8JB0300253B0H.html 11 revised Budget Law limits the rights of local governments to issue debts. Quota is applied to the issuance of subnational debts. The authority and purpose of local governments are strictly controlled and the subnational debts are included into the budget management and supervised by the People’s Congress. The government financing function is separated from the financing platform companies. The liabilities of government and enterprises shall be separated. It is prohibited from transferring the enterprise’s debts to the government. The local government debt risk assessment and pre-warning mechanism shall be established, as well as the emergency response mechanism and accountability system. The central government adopts a non-aid principle. The institutional framework for the government and social capital cooperation model has been established and related laws are also in drafting and formulating. The following efforts are needed to improve the local debt financing mechanism. We need to complete the short-term tasks, control currently expanding debt scale and eliminate the threats on China’s economic development, social stability and macro-management. Also, at the micro-level, we need to establish and improve debt scale, and risk and efficiency control system. At the macro-level, we need to make a coordinated reform in the aspects of economy, legal environment and financial system. In the short term, we urgently need to establish an accurate, comprehensive and timely local government debt statistical system, internal reporting system and external disclosure system. We need to improve local government debt issuance, repayment and management system, suitably expand local government bond scale, transfer annual quota management to balance management and strengthen the discretion of local government to issue and manage government bond. We will further regulate and innovate the local government financing mechanism, and orderly promote the cooperation between government and social capital. Some government assets will be revitalized through the securitization of assets. We actively establish the fund system for government capital expenditure (e.g. government guide fund and industry investment fund). In the aspect of medium and long-term, we need to actively play the role of legal regulation, rule control and market constraint. The boundary of government function and behavior shall be clearly distinguished. Through the legislation, we will regulate the administrative power, financial power, scope of revenues and expenditures and expenditure responsibilities of government at all levels. The local governments will be gradually granted controllable and flexible rights to borrow loans. The local government debt financing management system will be established jointly supervised by the central government, local authorities, people and market. Specific Suggestions In line with China’s current situation, to control the local government debt risks, we need to consider the following two aspects. On the one hand, we need to mitigate local government risks and ensure not to affect the economic growth. Only economic sustainability could ensure the financial sustainability. Thus, we need to expand investment and strengthen economic operation capacity so as to ensure to meet preset economic growth targets through increasing debts in a certain time. On the other hand, we need to improve government investment efficiency, control the growth rate of government debt and control financial risk. 12 Integrate investment and financing, and focus on investment efficiency To control risk and ensure financial sustainability is a core issue for investment and financing. Someone believes that the diversified financing channel is a source of local government debt risk. Actually, we need to re-examine whether bond financing could eliminate risks. There are always risks, even we solve the financing channel risks. If we consider financing and investment as a whole, the huge risks for the local government lie in investment rather than financing. If the investment is poor, risks are still bigger even borrowing less. If the investment is efficient, the risks could be mitigated even borrowing more. Previously, many local governments invested in new emerging strategic industries, such as photovoltaic industry. Such decision sounds alright for the local government. However, this strategy is wrong from the aspect of the whole country as the local government rushed to launch photovoltaic projects, leading to the bankruptcy of many photovoltaic enterprises. Therefore, it is important to innovate the investment mechanism and investment methods. We need to improve investment efficiency and transparency. It is much more important for regulating government investment behavior than regulating financing. Attract non-public investors to invest in public sectors by fully using the market-oriented mechanism Traditionally, people consider that the investment in public sector shall be borne by the government. However, public sector investment is still profitable. Now, we promote the PPP model (Public-Private Partnership), aiming to introduce the social capital into the public service sector. Originated from the UK in the 1970s, the PPP model has been widely applied in many countries across the world. Through sharing interests and risks, this model could not only effectively provide public service and mitigate government debt risks, but also provide stable profits for the social capital. In the long-term, the PPP model exhibits great potential. Of course, we need to legalize and regulate the PPP model and avoid borrowing loans in the name of PPP. In addition, we could use the PPP mechanism to establish the fund cooperation model for the government and social capital. Thus, we will actively establish the fund financing system for government capital expenditure (e.g. government guide fund and industry investment fund). We will suitably introduce social capital, expand the role of financial capital, and establish the new fund management mechanism featured as the market-oriented management and high Deepen budget reform and revitalize the financial stock fund Now, the problems we face are: on the one hand, the local government debt keeps growing with increasing risks; on the other hand, the local governments have a large amount of idle funds, which are mainly caused by fragmented budget and capital. The phenomenon of increasing idle capital while keeping borrowing is quite common. Thus, we need to solve this issue through budget management system. Accelerate government accounting reform and provide a scientific basis for identifying and swapping debt stock The current local government debt stock includes both government and corporate debts, which is mainly a result of lacking the standard of recognizing government debts. The acceleration of the government accounting reform is the only solution to solve this problem fundamentally. This is also a basis 13 to strengthen the foundation of local government debt management and identify the debt stock. In 2015, the swapping quota of RMB 3 trillion is allocated to help to swap the debt due this year. In the future, we need to accelerate the government accounting reform and establish the government debt standard to accurately identify the swapping scale of local government debt stock. Emphasize pre-warning management of local government debt risk and risk control plan On the whole, China’s local government debt risk is controllable. However, debt crisis may occur in some regions. Thus, we need to pay attention to the pre-warning management and risk identification for local government debt risks. The regional and systematic risks shall be earnestly prevented. Finally, liquidity risk is the main risk to be prevented for the prevention of government debt risks. Thus, we need to examine related indicators of subnational debts that are due, pressure for repayment of the principal and interest, re-financing space in the financial market, the establishment of redemption fund, the financial revenue capacity and imbalance between revenue and expenditure in the current term. We will design related management and control plan and hold the bottom lines to prevent regional and systematic crises in China. China's Subnational Debts: Problems and Suggestions Research Institute for Fiscal Science, Ministry of Finance Liu Shangxi 14 System Characteristics and Reform Logic of China’s Subnational Debts Qiao Baoyun System Characteristics and Reform Logic of China’s Subnational Debts Central University of Finance and Economics China Academy of Public Finance and Public Policy Qiao Baoyun 15 SECTION II - Looking at the Past through Country Experiences North America (USA) – “The United States: Regulatory Systems of Subnational Debt and Markets”. Dr. Lili Liu, World Bank, Governance Global Practice, EFI Vice-Presidency Emerging Markets: “Subnational Debt Management in Brazil and Mexico”. Dr. Fernando Blanco, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency Emerging Markets: “Subnational Debt: The Case of the Russian Federation”. Dr. John Litwack, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency Emerging Markets: “Subnational Debt Restructuring in India”. Ms. Abha Prasad, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency European Union (EU): “Fiscal Rules and Debt Restructuring” (Greece and Ireland). Dr. Christoph T. F. Ungerer, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency 16 The United States: Regulatory Systems of Subnational Debt and Markets Lili Liu The United States: Regulatory Systems of Subnational Debt and Markets Dr. Lili Liu Global Lead, Decentralization and Intergovernmental Finance Governance Global Practice Equitable Growth, Finance and Institutions Vice-Presidency The World Bank October 22, 2015 For further details see: Canuto, Otaviano; Liu, Lili. 2013. Until Debt Do Us Part: Subnational Debt, Insolvency, and Markets. Washington, DC: World Bank. © World Bank. License: CC BY 3.0 IGO. https://openknowledge.worldbank.org/handle/10986/12597 Chapters 8, 9 and 14. 17 Subnational Debt Management in Brazil and Mexico Fernando Blanco BRAZIL Subnational Debt Framework The fiscal behavior and indebtedness of state and municipal governments in Brazil have traditionally been a major source of macroeconomic instability in the nineties. Spending obligations derived from the decentralization fostered by the Constitution of 1988 and a profligate fiscal stance had been a common characteristic among Brazilian subnational governments through the latter half of the 1990s. Expansionary fiscal policies and the lack of effective indebtedness controls resulted in frequent subnational debt crises. This resulted in three debt-refinancing operations (bailouts) by the Federal Government to state government and 2 bailouts to municipalities between 1989 and 1997. The first subnational fiscal crisis and bailout took place in 1989, very soon after approval of the new Constitution. The Federal Government assumed part of the states’ external debt, which totaled 2 percent of GDP. It proved insufficient to address the state debt issue. In 1993, the Federal Government financed another subnational debt bailout, this time refinancing the debt of states and municipalities amounting to 7.2 percent of GDP, mostly with federal financing agencies. It also assumed some of the liabilities of the states and municipalities with other financial institutions and contractors. The Federal Government itself on some occasions restructured part of its liabilities to financial institutions, including international financial institutions. Following the Real Plan of 1994, price stability was achieved at the cost of high domestic interest rates which increased public sector indebtedness including the subnational one, making Brazil’s macroeconomic situation vulnerable to external shocks such as the Asian and Russian financial crises in 1997/98. This spurred the Federal Government to make a stronger effort to address the issue of subnational government indebtedness via a third comprehensive bailout in 1997 for states and in 2001 for municipalities with the National Treasury (STN) simply assuming the bond debt of state governments and state financial institutions. The terms for the debt restructuring were as follows: repayment period at 30 years, interest rate at 6 percent plus inflation for most states and 7.5 percent plus inflation for a few states, debt servicing on the refinanced debt, the so-called “intra-limite” debt, was capped at 13 percent of a state’s net revenues (Receita Liquida Real - RLR), defined as the current revenue, net of transfers to municipalities and earmarked revenues. Any debt service above the 13 percent cap was deemed residual, and automatically recapitalized and added to the intra-limite debt stock. The accumulated residual debt (debt service above the 13% ceiling) at the end of the contracts (in 2028 for states and 2032 for municipalities) were to be paid off in 10 years. Unlike the two previous refinancing operations, the 1997/2001bailout was conditional. Each of the 25 states that had their debt rescheduled needed to sign a debt negotiation contract with STN, agreeing to comply with a fiscal adjustment and structural reform program. These were specified in three-year rolling 18 Fiscal Adjustment Programs (Programas de Ajuste Fiscal – PAFs), which set annual targets or limits on gross indebtedness, primary balances, personnel spending, own tax revenue collection, credit operations, guarantees, and public investment, in order to assure a gradual decline in indebtedness. An important objective was to ensure that the gross debt did not exceed net real revenues by 2028. The PAFs also include structural reforms such as those relating to privatization and public sector modernization initiatives. The controls on subnational fiscal performance were further strengthened by the approval of the Fiscal Responsibility Law (Lei de Responsabilidade Fiscal – LRF) in 2000. The LRF institutionalized fiscal discipline at all levels of government, incorporating hard budget constraints into a single unifying framework. It explicitly prohibits debt refinancing operations between different levels of government, thereby addressing the moral hazard problem in intergovernmental fiscal relations caused by sequential bailouts. Complementary Senate resolutions also prohibit borrowing if: (i) the net consolidated debt exceeds twice net current revenue (RLC – Receita Liquida Corrente); 3 (ii) new credit operations exceed 16 percent of RLC, and (iii) debt service exceeds 11.5 percent of RLC. Borrowing is also prohibited if it violates the debt reduction schedules set by the debt renegotiation contracts under the Law 9496. Finally, emission of subnational governments bonds is generally prohibited through 2016; however, states whose net debt is less than net current revenue can issue bonds after 2011, although even here the Federal Government retains to the option to review the decision to issue bonds. On top of these regulations, supply side constraints were also adopted. Subnational overall ceilings for credit operations by federal financial institutions and domestic banks were established by a series of Resolutions by the National Monetary Council (CMN) between 2001 and 2006. This further restrained the access to credit by subnational governments as the ceilings were very low (around USS2 billion) for the entire subnational sector. Strict observance of debt renegotiation contracts and the LRF have resulted in robust primary surpluses in the state governments. These surpluses, which represented about 25 percent of the overall fiscal adjustment effort of the Brazilian public sector, have been key to the improvement of the country’s macroeconomic conditions. The prudent fiscal stance adopted by both states and municipalities, combined with improved revenue growth performance, resulted in a significant decline in subnational indebtedness and a pronounced improvement in subnational indebtedness indicators since 2003. This system of controls has resulted in a substantial macroeconomic adjustment as evidenced by repeated state and municipal surpluses—an adjustment that has continued through four federal administrations and two presidents. After ten years of consolidating macroeconomic credibility, it is useful to assess how the system has stood up under time and responded to changing domestic and international circumstances and to consider how it might further be developed. Subnational debt fell from 18 percent of GDP in 2003 to 10 percent of GDP in 2010. 3 The net real revenue (RLR) is defined as the total revenue less credit operations, asset sales, capital transfers, transfers to municipalities, FUNDEF contributions, and the social security contributions of civil servants that opt out of the state system. RLR is smaller than the RLC used in the PAF. Also, the debt contracts cover only the central state governments while the LRF includes the consolidated state public sector. In the same vein, the debt renegotiation contracts use gross debt for state indebtedness while the LRF uses the net consolidated debt. 19 While a success overall, the controls embodied in the Fiscal Adjustment Programs (PAFs) and Fiscal Responsibility Law (LRF) have also had the following consequences:  The cap on debt service payments has led to a large build-up of capitalized interest by some major states and municipalities. Some of this debt may not be able to be paid when it comes due in 2028 and 2032, which could lead to pressures for another bailout. They are already affecting the credit risks of some states as well as the Federal Government.  The debt owed by states and municipalities to the Federal Government accounts for the bulk of the National Treasury’s assets and is an asset that the market values below its face value.  Debt-service payments were heavily front-loaded for states with a low indebtedness level: for some of these states, projected payments in 2009 are roughly three times those in 2028, implying an uneven distribution of the debt burden. At the same time, the system of controls does not allow states to borrow in order to smooth out the tax burden or public investment stream. In some years there is no money for investment. In the future years, there will be excess of resources for investment.  States and municipalities have exerted strong pressure on the Federal Government to ease up on restrictions via ad hoc measures. In 2010-14 and as part of the countercyclical policies adoped to address the global financial crisis, there has been a strong relaxation of the fiscal rules that were very costly to the Federal Government and sent the wrong signal on how best to manage subnational finances. The risks of political pressures for the adoption of ad hoc opaque measures materialized. As a result, without the previous hard budget constraints, subnational governments public finances strongly deteriorated and their debts interrupted their declining trend. MEXICO Subnational Debt Framework Supporting a market approach Mexico’s subnational debt framework was substantially reformed in 2000. This reform have fostered the expansion of a domestic credit market for state and municipalities. The previous framework was based on the concept of the mandato (mandates) which consisted in the federal government acting as a trustee in servicing subnational debts that had been collateralized with the unconditional revenue-sharing transfers known as participaciones. In practice, the mandato was perceived by the markets as a guarantee by the federal government of subnational debt. This perception of a very likely federal bailout created two problems: (a) banks had the incentive to provide lending to subnational governments without assessing the borrower’s repayment capacity since they perceived them to be risk free; and (b) subnational governments also had the expectation of a bailout since it was not credible that the federal government would in fact reduce transfers in case of a debt crisis. The reforms regarding the new regulatory framework for subnational debt were based on two main concepts: an explicit no-bail-out commitment by the federal government and a new system aimed at 20 enabling lenders to correctly assess idiosyncratic subnational risks. These objectives were pursued through (a) the elimination of the mandatos and the creation of subnational governments’ Master Trust Funds; (b) establishment of a link between the capital risk weighting of bank loans to subnational governments and their credit rating; (c) and a requirement to register subnational loans with the Ministry of Finance, conditional on being current on financial transparency requirements. The establishment of the Master Trust Fund (MTF) was a key factor to reduce risks and borrowing costs. In March 2000, the federal government approved a Master Trust contract that regulated and established payment procedures for debt issuances guaranteed by tax participations. Each subnational government would establish its own master trust, adjusted to its specific legal environment. With the enactment of this regulation and the creation of the master trust instrument, the central government unveiled the first stage of development for subnational structured financing. The States transfer the flow of Participaciones to the master trust through an irrevocable instruction to the federal treasury, while the municipalities transfer their tax participation flow through an irrevocable instruction to their state treasury. There is an irrevocable transfer of present and future tax participations, which facilitates the legal analysis of the securitized debt, as it allows the isolation of the payment source from the issuer. The legal analysis is crucial to assess the probability that a state government will try to deviate flows to cover expenses rather than the debt service of the issuances. Upon the establishment of the MTF, the legal and financial frameworks for subnational indebtedness became an important aspect of the borrowing process. Some subnational governments also introduced reforms in their debt laws and fiscal codes to meet the legal requirements harmonizing budgeting, financial management, fiscal codes and debt regulations for the three levels. Supply side regulations were also strengthened. In August 2004, the National Banking and Securities Commission (CNBV) introduced changes to the regulations for commercial and development banks on loans to subnational entities. Loans granted to SNGs with outstanding amounts equivalent to or above certain threshold had to be provisioned according to the risk level determined by the credit rating assigned to the SNG by at least one external rating agency. If the SNG was rated by two rating agencies, the lowest rating is the one considered. The new provisioning rule also took into consideration loans that benefit from a credit enhancement mechanism provided by credit or financial assets, such as by the pledging of tax participation revenues. The CNBV risk level was also adjusted depending on the credit, financial, and legal strength of the structure or the guarantee mechanisms for the loan, and penalized the loan or the underlying structured transaction by assigning a high risk level if the indebtedness was rated by only one agency. In October 2011, the provisioning regulations were changed. Loans to SNGs no longer have to be provisioned according to credit ratings but on the basis of expected losses, which depend on the payment history of each entity and its present financial situation. Still, debt of public sector entities represents a high share of private banks’ assets. Eight out of the ten larger debtors in the domestic credit market are public sector institutions (6 states and 2 parastatals). Moreover, there are some banking institutions with loans that are extremely concentrated in the states and 3 banks with loans to only one entity that are the equivalent of their base capital. It must be noted, 21 however, that the concentration of loans is by amount and does not consider the credit quality or the guarantees of the financial structures. Credit ratings also reinforced the subnational indebtedness framework based on market discipline. The use of credit ratings was encouraged by means of penalizing the reserve requirements of nonrated borrowers, giving SNGs added incentives to keep a record of their finances. By 2010, all states and a growing number of municipalities had been assigned credit ratings by at least one recognized rating agency. As a result, SN public finances are subject to growing surveillance and scrutiny by private markets and rating agencies. Despite this link being modified in 2011, credit ratings are still a factor in determining the cost of borrowing for SNGs. The reforms described above have consolidated a market based approach for subnational lending. This approach assumes that financial markets exert fiscal discipline deterring excessive indebtedness through higher interest rates and exclusion from credit markets. For market-based fiscal discipline to work, adequate information on the borrower’s outstanding debt and repayment capacity should be available to potential lenders and there should be no perceived chance of a bailout by the central government. One of the main weakness of Mexico’s subnational indebtedness framework is the lack of reliable and uniform financial information. While Mexico adopted the General Governmental Accounting Law (LGCG) in 2007, its implementation at the subnational level has been limited. The LGCG establishes general criteria regarding the public sector accounts and the reporting of financial information of the public entities with a view of achieving harmonization. The Law also establishes a Council (Consejo Nacional de Armonización Contable) in charge of issuing the accounting norms and guidelines for the preparation of financial information to be used by the public entities, including States and municipalities. However, a large number of state and the majority of municipal governments have not adopted the uniform accounting methodology for SNGs finances set in the LGCG. The draft bill for SNGs fiscal discipline law mentioned above establishes requirements on establishing a sole Public Registry on debt which will strengthen the quality and uniformity of available information on the indebtedness of these governments. Moreover, compliance with this requirement will be a pre-condition to access federal guarantee on state debt which is introduced by the draft bill. Recent Initiatives to Enhance the Subnational Indebtedness Framework The possibility of debt restructuring operations conditioned to fiscal adjustment agreements has been recently incorporated into the overall subnational indebtedness framework. Since 2013, seven state governments have signed debt-restructuring operations with private banks conditioned to fiscal adjustment plans agreed with the Ministry of Finance. As the existing indebtedness framework does not encompass the participation of the federal government in the agreements, public banks participate in these agreements as an indirect way in which the Ministry of Finance can influence the definition of fiscal targets to the debt restructuring operations. These agreements seek to strengthen the fiscal position of the states and reduce debt-restructuring costs. As a result of the increased indebtedness in some states, the federal governments has decided to enhance the existing indebtedness framework by establishing hierarchical controls on subnational fiscal 22 performance and indebtedness. In February 2015, the National Congress approved a Constitutional Amendment that was ratified by the majority of states Assemblies and came into effect in May 2015, which grants authority to the federal level to enact laws regulating the finances and indebtedness of the subnational governments (SNGs). Prior to the Amendment, the National Congress did not have the authority to restrict or impose rules on the SNGs on contracting debt as they were sovereign in this respect. In exchange for relinquishing this sovereignty, the SNGs will have the possibility of obtaining federal government guarantees on their debt restructuring operations conditional on the acceptance of their fiscal adjustment plans by the Ministry of Finance. This will provide the legal framework for the participation of the federal government in the debt restructuring operations mentioned in the paragraph above. Guidelines on registration of existing obligations have already been issued. The main changes introduced by the Amendment are:  Fiscal responsibility. The amendment grants the authority to the national Congress to enact laws on fiscal responsibility and fiscal coordination; define the conditions for federal guarantees on debt and restructuring or refinancing operations; regulate the indebtedness of the SNGs, set limits and define modalities through which SNGs are permitted to earmark participaciones as collateral for financial obligations, require SNGs to register and publish all their loans and obligations in a Public Registry, in a timely and transparent basis; define an alert system regarding the debt management operations of the SNGs; define the sanctions to be applied to civil servants that do not comply with these regulations; establish a bicameral commission to examine the fiscal adjustment agreements; and enact laws based on the stability principle.  Federal Guarantee of SNGs debt. The amendment grants the authority to the National Congress to define the conditions for federal guarantees on debt and restructuring or refinancing operations.  Monitoring SNGs debt. The Court of Accounts (Auditorí a Superior de la Federación – ASF) was given the authority to monitor SNGs debt, the federal guarantees on SNGs borrowing and the use of these loans, and to audit compliance with federal programs’ objectives.  Responsibility of civil servants. Establishes responsibility for the civil servants in charge of the management of the SNGs’ public resources and debt. This measure will be reinforced by the enactment of the Anti-Corruption Law.  Local legislatures. It adds to the responsibilities of the local legislatures the monitoring of the SNGs actions in regards to funds, local resources and public debt.  Conditions for borrowing. Other than for productive investment, loans may be used for refinancing or restructuring of debt, and in the case of the states to provide guarantees on loans to municipalities, but never for current expenditure. Borrowing must be effected under best market conditions. Local legislatures are required to approve, by two-thirds majority, the ceilings for indebtedness, analyzing the purpose of the loans, payment capacity, the guarantee that is provided and the source of payment. SNGs may contract loans to cover short-term needs, without exceeding the limits and conditions of the Public Debt Law enacted by Congress. Short-term 23 obligations must be repaid at least three months before the end of the SNG administration and no new obligations may be contracted during this three-month period. The next step regarding the fiscal and financial discipline of the SNGs will be the approval and implementation of a Fiscal Responsibility Law for SNGs. In line with the Constitutional Amendment enacted in May, a draft Bill on SNGs finances (Ley de Disciplina Financiera de las Entidades Federativas y los Municipios - LDFEFM) was submitted to Congress in August 2015 and it is expect to be approved in December 2015 to begin to be applied since 2016. The LDFEFM contains rules on: (i) fiscal balances and other indicators, procedural and numerical, and escape clauses; (ii) budgetary norms; (iii) contracting of debt and other obligations, including short-term obligations, and on the use of participaciones as collateral for borrowing; (iv) the role of the local legislatures; (v) conditions on the fiscal adjustment agreements; (v) defining an alert system on debt management by SNGs; (vi) requirements to register debt and other obligations in a Public Registry; (vii) sanctions to civil servants that do not comply with the Law; and (viii) conditions to provide federal government guarantees on borrowing by SNGs. Subnational Debt Management in Brazil and Mexico Fernando Blanco Lead Economist Macroeconomics & Fiscal Management Global Practice Equitable Growth, Finance and Institutions Vice-Presidency The World Bank Presentation at Ministry of Finance, P.R.China-World Bank Summit on Subnational Debt Management and Restructuring, Nanning, Guangxi Province, P.R. China. October 22, 2015. 24 Subnational Debt: The Case of the Russian Federation John Litwack Background The Russian Federation consists of 85 Subjects of the Federation and thousands of municipalities and districts. The fiscal federalist system in the Russia has evolved considerably since the breakup of the Soviet Union, although managing and controlling subnational debt continues to be a challenge. In more recent years, this problem reflects a growing imbalance between subnational revenues and expenditure obligations in a number of Russian regions. There is a very high degree of formal centralization in the determination of subnational revenues and expenditures in the Russian Federation. Over 85% of revenues at the subnational level consist of revenue sharing from taxes that are set or tightly regulated by the Federal Government. Similarly, a high share of expenditures at the subnational level correspond to mandates from Federal legislation or instructions. However, regional and local governments in Russia have had a significant amount of largely informal fiscal authority on their territories. The strong regulatory authority of regional administrations has underpinned various forms of partnerships with firms and economic organizations for the conduct of quasi-fiscal activities. The informal authority of subnational administrations was particularly strong before 2005. Decisions in the second half of 2004 that changed the rules for election to Parliament and allowed for the removal of regional governors by the President fundamentally altered the balance of power from regions to the center, and particularly to the presidency. The Russian experience in subnational debt The story of subnational debt in the Russian Federation can be divided usefully into three different periods:  1991-1998: During the 1990s, Russian regional administrations (Subjects of the Federation) had exceptionally large (mostly informal) authority on their territories. They were free to borrow under few constraints, and this was formalized in a Budget Law of 1993. Subnational debt in Russia accumulated rapidly, followed by the virtual insolvency of a large number of regions during the financial crisis of 1998-1999.  1998-2000: In this difficult period, most Russian regions were severely liquidity constrained, with past debt obligations in arrears. In this context, regional budgets became demonetized. By 1999, the majority of revenues and expenditures in the budgets of most Subjects of the Federation were in money surrogates such as barter, bills of exchange (veksels), or tax offsets.  2000-2008: Until the oil price shock of late 2008, most Russian regions experienced steady improvements in their financial situation. Budgets again became monetized, past debts were worked out, mostly in a decentralized manner, and subnational debt fell steadily to under 2% of GDP. These improvements were facilitated by the stabilization of the macroeconomic situation in Russia, higher oil prices, economic growth, and budgetary reforms discussed below. 25  2009-2015: Since 2009, Russia has once again faced the challenge of weaker oil prices, as well as the recent economic sanctions. Subnational revenues have been much lower than expected, reflecting slower growth, while expenditure obligations have been increasing rapidly. Subnational debt has steadily increased to 3.4% of GDP, a number of oblasts are experiencing serious difficulties, and loans from the Federal Government to Subjects of the Federation to relieve financial distress have grown significantly. A statistical assessment of subnational debt in Russia in the late 1990s is difficult, as comprehensive data were not available at the time. A first relatively comprehensive survey was conducted in January, 1999, but limited to 53 Subjects of the Federation. It revealed that the explicit debt these regions had reached 8.2% of GDP. This does not account for substantial additional expenditure arrears. Thirty five percent of the explicit debt was delinquent (in arrears). There was no comprehensive bailout at the time from the Federal Government, which was also in default in light of the crisis. But selective financial support in the form of federal loans was extended to some regions. Loans from the federal government, mostly to restructure debts and reduce arrears, reached 17% of subnational debt in these 53 regions. The Budget Code was first adopted in 2000, and has subsequently experienced some major revisions and amendments. The enforcement of the Budget Code and other fiscal reforms proceeded in the context of political changes that shifted much more power to Presidency and Federal Government. Some of the most important components of the Budget Code introduced during 2000-2004 include:  A major effort to better define and delineate assignments of revenues and expenditure responsibilities between budgets.  Strict limits on external borrowing: only regions with credit ratings from at least two major agencies at the same level as that for the Federal Government are eligible to borrow externally.  Limits on domestic borrowing by regions: (a) the stock of debt cannot exceed 100% of annual revenues (50% for regions highly dependent on transfers), (b) debt servicing cannot exceed 15% of new expenditures. Regions in violation of these conditions cannot issue new debt.  Comprehensive budget accounting required, including for all subnational debt and guarantees.  Equalization transfer formula made more stable (to combat soft budget constraints).  No tax revenues can be received in money surrogates.  Temporary financial management by higher level of government can be imposed on insolvent subnational administrations. This concerns regions with arrears on debt service exceeded 30% of own revenues. The period of 2000-2008 witnessed steady improvements in the financial situation of most Russian regions. Higher oil prices and the first significant economic growth in Russia were important factors behind the recovery, but the provisions of the Budget Code arguably also played a role. Russian regions worked out or restructured their past debts in a largely decentralized manner, although the Federal 26 Government gave selective assistance in some cases. By 2008, subnational debt and guarantees in Russia fell to less than 2% of GDP. Subnational budgets again became monetized. Subnational Debt and Gurantees in Russia (Share of GDP) 4.00 3.00 2.00 1.00 0.00 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 + As illustrated in the figure above, the oil price shocks of 2009 and 2014 can be associated with new increases in Russian subnational debt, which reached 3.4% of GDP by 2015. More favorable economic conditions in 2012-2013 again contributed to a decline in subnational debt as a share of GDP, but debt has risen at a fast pace in 2014-2015 in the context of the oil price shock, much slower growth (currently recession), international economic sanctions, and higher expenditure obligations for the regions. Significantly higher expenditure obligations in the social sphere can be traced to a serious of decrees and instructions following the 2012 elections that called for greater financial commitments to heath, education, and other social expenditures. A large share of this burden fell on regional budgets. On aggregate, subnational debt as a share of subnational own revenues (net of federal transfers) still stood at only 34% in early 2015. But this aggregate figure masks huge differences between oblasts. In early 2015, almost half (40) the Subjects of the Russian Federation reportedly had debts exceeding 60% of own revenues. Furthermore, ten Subjects of the Federation are reportedly close to the limit specified in the Budget Code of debt at 100% of own revenues. Subnational Debt and Guarantees in Russia (Share of Own Revenues%) 40 30 20 10 0 2011 2012 2013 2014 2015 27 The below figure, which shows the evolution of the composition of Russian subnational debt, is revealing of the current difficult situation. It shows that the recent growth in debt at the subnational level in Russia has become increasingly concentrated in commercial bank credit and loans from the Federal Government. Commercial bank loans are generally short-term and quite costly, implying an unfavorable trend in the maturity structure and cost of subnational debt. Federal loans are granted largely as financial assistance to more distressed regions that are struggling with debt servicing. The Federal Government has also waived counting of federal loans to regions in calculating the debt ratio requirement (100% or 50% of own revenues) until 2017. As of January 1, 2015, bank credit accounted for 46% of subnational debt, while federal loans accounted for 33%. The Composition of Russian Subnational Debt 2500.0 (bn rubles) 2000.0 1500.0 1000.0 500.0 0.0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Securities Bank Credit Guarantees Foreign Debt Credit From Federal Budget Some lessons from the Russian experience The experience in Russia with subnational debt has been unique in some ways, given peculiarities of the political, institutional, and economic situation during the last 24 years of building a market economy. Nevertheless, some general lessons can be highlighted that are relevant for other countries as well. 1) Recent pressures for the accumulation of subnational debt in Russia reflect a growing imbalance between subnational revenues and expenditure obligations. General balance between revenues and expenditure obligations is key to controlling the build-up of subnational debt. 2) The threat of insolvency and temporary administration by a higher level of government has lacked credibility in Russia. Since Russian regions have little control over their own revenues and expenditure obligations, it is difficult to hold them responsible for controlling growth in debt . 3) Subnational governments in Russia have primary responsibilities for social expenditures during a time of slower growth and lower revenues. Maintaining control of subnational debt in a time of changing economic circumstances can require fiscal adjustment. 28 John Litwack Lead Economist World Bank Prepared for P.R. Ministry of Finance-World Bank High-Level Summit on Subnational Debt Management and Restructuring, October 22, 2015, Nanning, Guangxi Province, P.R. China 29 Subnational Debt Restructuring in India Abha Prasad The buildup of Sub-national Debt There have been three major phases in the evolution of fiscal indicators for subnationals in India. The first episode, pre-1988 was one of sustainable debt and stable fiscal indicators. The second episode, which lasted from FY2000 through FY2003 and may be deemed “fiscal slippage,” was characterized by high nominal interest rates, low nominal rates of growth, and a rise in public sector remunerative expenses following the recommendations of the Fifth Pay Commission in 1998.4 The third episode, which may be termed “fiscal consolidation,” began in FY2004 after the central government adopted the Fiscal Responsibility and Budget Management Act (FRBMA) in 2003 and the recommendations of the 12th Finance Commission (FC) incentivized adoption of similar legislations at the State5 level in 2004. The period of fiscal consolidation was enforced with legislative changes introduced at the national and subnational level. The central government enacted a Fiscal Responsibility and Budget Management Act in 2003 to ensure medium to long term fiscal sustainability of central finances, by imposing limits on borrowings, debts and deficits. Fiscal consolidation by subnational governments was incentivized following recommendations of the 12th Finance Commission.6 Recommendations of the 12th FC, for the period 2005-2010, were released in 2004 and accepted in full by the government, proposed a debt restructuring plan for the subnational governments wherein the proposed debt relief was explicitly linked to rules-based legislative reforms. This was landmark in advancing fiscal reforms at the sub-national level by linking availability of debt relief7 for central loans contingent upon the enactment of rule-based fiscal responsibility legislations. All states were eligible to obtain relief from the year they enacted fiscal responsibility legislation. Key features of the 12th FC recommendations include: 8  Subnationals should be allowed to consolidate and reschedule their outstanding loans to the Centre (with a lower-than-market interest rate of 7.5 percent). The scheme was subject to enactment of Fiscal Responsibility Legislations (FRLs), which prescribe specific annual fiscal targets, elimination of current deficits by 2008-09, reduce fiscal deficits to 3 percent, and setting 4 FY2000 refers to fiscal year ending March 31, 2001. 5 States refer to state governments, also used for subnationals here. 6 As per article 280 of the Indian Constitution, a union finance commission is constituted every five years to make recommendations on resource transfers between different levels of the government. Revenue sharing is based on a formula-based approach of the Finance Commission (FC) to allocate taxes and grants, with the objective of filling the expenditure-revenue gap (deficit financing). The revenue sharing formula is motivated by the concepts of vertical and horizontal equity: (i) vertical sharing between the centre and States whereby all central taxes and excise duties are combined in a divisible pool of central taxes; (ii) horizontal sharing among States whereby the FC attempts to correct the differentials in revenue capacity and cost factors inherent in the diversity of States. 7 States have been provided debt relief since 2nd FC for loans extended by Centre with a view to aiding their debt sustainability. These included (i) debt consolidation on common terms and reduction of interest rates, (ii) rescheduling of loans to elongate repayment without changes in interest rates, (iii) moratorium on interest and principal repayments for certain period, (iv) debt write-offs and (v) introduction of schemes that linked debt relief to fiscal performance. 8 Draws from Rangarajan and Prasad, World Bank (2012). 30 up of guarantee redemption funds through earmarked guarantee fees, to meet contingent liabilities.  Subnationals should be allowed to approach the market directly for borrowings. If some fiscally weak Subnationals are unable to raise funds, then they could borrow through on-lending from the Centre. Institutional reforms To implement 12th FC recommendations, a Debt Consolidation and Relief Facility was introduced in 2005–06, which provided debt relief through consolidation, rescheduling repayments for a fresh term of 20 years, and lowering of interest rate on the debt to 7.5 percent. Fiscal correction was given an impetus with introduction of a ‘debt swap scheme’ during 2003-04 to lower the existing interest burden and increase market access. Loans from the Centre (with higher interest rates) were substituted with market loans and small savings proceeds (at lower rates). In addition, repayments for 2005-10 on central loans contracted up to March 31, 2004 (after consolidation and rescheduling) were eligible for write-off subject to reduction in revenue deficits. This was also subject to containment of the fiscal deficit to the 2004–05 level. Further, if the revenue deficit was brought down to zero by 2008 –09, the entire repayments during 2005–10 were written off. In conjunction with state debt relief, the 12th FC also enforced that for future lending the central government will not act as an intermediary, and states were to approach markets directly (domestic). For external loans, same terms were passed on directly to states. Outcomes are encouraging Following the adoption of the recommendations of the 12th FC, all Subnationals (except, Sikkim and West Bengal) enacted their FRLs – which specified targets for deficits and debt ratios over the medium term, but varied significantly across states in terms of time of enactment, targets for fiscal consolidation, provisions for contingent liabilities and debt sustainability. Subsequently there was variation in terms of the states’ fiscal performance. Table 1: Change in Subnationals’ fiscal performance between 1998 to 2003 and 2004 to 2009 Bihar, Chhattisgarh, Haryana, Karnataka, Improved fiscal performance Madhya Pradesh, Orissa, Punjab and Uttar Pradesh Goa, Jharkhand, Kerala, Maharashtra, Continued fiscal deterioration Uttarakhand and West Bengal Source: Rangarajan and Prasad, World Bank (2012). Performance is assessed against the median performance of General Category States on primary deficit, during the specified time periods. The impact from enactment of FRLs has been overall positive, as can be seen by the recent Reserve Bank of India study on State Finances (Figure 1). The study analyzes a set of fiscal deficit indicators of Indian states, comparing mean indicators recorded pre-FRL (1992-93) and post FRL (2012-13) period. Selected findings from this analysis follow below. 31  Revenue Deficit: Before FRL, revenue accounts of all 17 non-special category (NSC) states were in deficit; special category states had marginal surpluses in the revenue account. After FRLs, 12 of the NSC states recorded revenue surpluses, including all low-income states. SC states recorded larger surpluses post-FRBM due to large central transfers.  Gross Fiscal Deficit (GFD): Before FRL, 15 out of 17 NSC states GFD was above 3% of their gross state domestic product (GSDP) on average. After FRL, all group A states and 6 out of 7 group C states lowered their GFD/GSDP to below 3%.  Primary Deficit: Group C states recorded highest primary deficit pre-FRL. After FRL, primary deficits of all states came down; Group C states recorded primary surpluses. Figure 1. Deficit indicators of states: Mean of pre- and post-FRLs Only for non-special category states: Group ‘A’ constitutes the top five states in terms of their real per capita incomes as in 2013-14. It includes Goa, Maharashtra, Haryana, Gujarat and Tamil Nadu. Group ‘B’ constitutes the middle income states viz., Kerala, Punjab, Karnataka, Andhra Pradesh and West Bengal Group ‘C’ includes remaining Rajasthan, Jharkhand, Chhattisgarh, Madhya Pradesh, Odisha, Uttar Pradesh and Bihar. ------------------------------------------------------ SC = Special Category states: Arunachal Pradesh; Assam; Himachal Pradesh; Jammu and Kashmir; Manipur; Meghalaya; Mizoram; Nagaland; Sikkim; Tripura; Uttarakhand -------------------------------------------------------- Source: Reserve Bank of India, 2015, State Finances: A Study of Budgets of Pre-FRL period: 1992-93 2014-2015. Post FRL period: 2012-13 One of the outcomes of fiscal responsibility measures adopted in India in early 2000s was the change in financing patterns with a move towards transparent market-based borrowings compared with the earlier dependence on loans from the Centre (Figure 2). In 1990-95, the share of subnationals’ market borrowings in total comprised 15%, while loans from the Centre comprised 47% of total. By 2013-14, market borrowings comprised 71% of total, while loans from the Centre comprised under 3% of total. 32 Figure 2. Financing of gross fiscal deficit Source: Reserve Bank of India, 2015, State Finances: A Study of Budgets of 2014-2015. On the Constitutional side, the 13th FC carried forward the momentum to support states toward fiscal correction and recommended return to the path of fiscal consolidation, following the global financial crisis. The recent 14th FC continued enforcing fiscal discipline: it recommended states’ ceiling on gross fiscal (overall) deficit of 3% of GSDP. Following the carrot and stick approach, additional borrowing limit of 0.25% is given if (i) debt/GSDP ratio is < 25% and/or (ii) interest payments/revenue receipts < 10%. And this additional borrowing is conditional on the state having ‘zero’ revenue deficit in the year in which borrowing is fixed and in the preceding year. Further, states were allowed to reset outstanding loans from the National Small Savings Funds, to a lower interest rate of 9 percent. Conclusion In summary the debt restructuring process in India adopted a two-pronged approach with incentives and regulation. Debt relief and debt restructuring efforts were given to states as incentives if the state governments enacted fiscal rules or set in place a credible path towards fiscal consolidation. On the debt restructuring side, it was recognized that States’ debt burden could not be alleviated unless reforms targeted both the: (i) overhang of high cost outstanding debt and (ii) flow of new borrowing at lower rates. These efforts at debt relief and consolidation were superimposed with the move towards market-based borrowings at lower interest rates. This had the effect of allowing States to reduce their dependence on loans from the Centre while moving to meet their financing requirements directly from market sources at market rates (Figure 2). This brought in the importance of market ‘watch’ and discipline in the system. It is also important to realize that the states moved gradually to access the market completely to meet their funding needs; a process almost across 10 years. In addition, to ensure sustainability of the process and avoid moral hazard of continued debt relief, states enacted fiscal responsibility legislations (FRLs) and put in place strong governance arrangements. Most states passed rules that targeted bringing the revenue or current deficit to zero and ensuring that all borrowings were taken only to meet capital expenditures. In addition debt was kept within bounds and caps instituted on issuance of guarantees, along with setting in place guarantee redemption funds. Evidence is clear that post-FRLs states lowered debts, interest payments and improved repayment 33 capacity (assessed as interest payments ratio of revenue receipts, Figure 3). This was partly as greater credibility was built and partly lower global interest rates that enabled states to borrow at cheaper rates. Figure 3. Debt, deficit and interest payments/revenue receipts Source: Reserve Bank of India, 2015, State Finances: A Study of Budgets of 2014-2015. In hindsight, a key lesson for other countries is the importance of incentivizing debt relief with prudent regulation or conditions that ensure the sustainability of reforms. Debt relief in isolation, without fiscal consolidation and legislative measures may not result in medium term debt sustainability. Finally, access to resources through market based mechanisms enables better outcomes than depending on central loans, on-lending or transfers. India: Restructuring Subnational Debt Abha Prasad Lead Debt Specialist Macroeconomics and Fiscal Management Global Practice, World Bank Group Presentation at the Ministry of Finance, P.R.China-World Bank Summit on Subnational Debt Management and Restructuring, Nanning, Guangxi Province, P.R. China, October 22, 2015. 34 Fiscal Rules and Debt Restructuring: The EU Experience Country Cases for Ireland and Greece Christoph T.F. Ungerer and Lili Liu This case study outlines a high-level overview of the original European Union (EU) fiscal framework in the 2000s. It proposes reasons why this framework did not contain the accumulation of fiscal, macroeconomic and financial imbalances in the European periphery during this period. Using the country cases of Ireland and Greece, this brief then discusses the economic adjustment programs that were implemented during the subsequent Eurozone sovereign debt crisis to help affected countries unwind these imbalances. It discusses possible reasons for why Ireland was able to exit its program after three years, while the Greek adjustment remains ongoing and is now in its sixth year. The brief concludes with a short overview of the EU fiscal framework reform since the crisis. Based on the EU experience, this brief draws lessons for the effective implementation of a fiscal framework in a federation of States as well as for the implementation of adjustment programs if members of the federation find themselves in fiscal difficulties. Country Context & Initial Regulatory Framework (Slide 2) In analogy to a subnational fiscal framework, the EU can be considered as a federation of 28 Member States (MS). This comparison is particularly applicable to the 19 MS of the Eurozone, which share a common currency. Nonetheless, relative to typical subnational entities, EU MS retain substantial sovereignty. For instance, the EU budget is only 1% of EU GDP. The EU fiscal framework is rules-based, in that the Maastricht Treaty and subsequent Stability & Growth Pact (SGP) put in place a system that obliges MS to enter Excessive Deficit Procedures (EDP) and take corrective actions if their fiscal stance breaches certain target values. Key metrics in this system are that MS should limit fiscal deficits to less than 3% of GDP and public debt to less than 60% of GDP. MS budgets are under surveillance by the European Commission (EC). Non-compliance with EDP procedures can lead to fines. In addition, to limit the liability of individual MS (and aiming to preserve market discipline on MS borrowing), the EU treaties contain a “no-bail-out” clause.9 Borrowing Discipline and the EU Fiscal Framework in the 2000s (Slide 3) During the 2000s, market discipline on MS borrowing was largely absent. Irrespective of fiscal stance and strength of the economy, countries such as Greece, Ireland and Germany faced largely similar long- term borrowing rates. This brief highlights three drivers of this breakdown in market discipline. First, the no-bail-out commitment of EU MS, particularly for Eurozone MS, was perceived as not credible. Second, 9 “The Union shall not be liable for or assume the commitments of central governments, regional, local or other public authorities, other bodies governed by public law, or public undertakings of any Member State” (TFEU. Art 125). 35 loose monetary policy conditions in the 2000s, coupled with a long period of relatively low volatility (the “Great Moderation”), led investors to insufficiently discriminate between investment opportunities. Third, EU banks faced undifferentiated risk-weights across MS sovereign bonds – they were not required to put up significantly more capital against sovereign bond issues of MS with stronger or weaker finances. In turn, this reduced incentives for banks to differentiate between MS bonds. In addition, the SGP rules suffered from implementation issues. First, there were issues in the reporting of fiscal statistics. This meant that budgetary weaknesses could build up, without resulting in full corrective action under the SGP. For instance, a new Greek government was forced to revise up the expected 2009 fiscal deficit from 4% to 12.5% of GDP in October 2009. Second, macroeconomic and financial imbalances hid underlying fiscal weakness. For example, the Irish banking system built significant exposure to a local property bubble in the 2000s. As this bubble burst in 2008, the Irish government was forced to inject capital in domestic banks, driving up public debt. A fiscal framework that focused on headline fiscal statistics did not adequately capture these spillover risks. Third, breaches of the SGP rules were insufficiently penalized. For many MS, being under EDP became the norm. A recent IMF study concludes that “about half of the countries have missed the 60 percent debt ceiling more than half of the time” between 1999 and 2013.10 The SGP came under particular strain, when its biggest members and supporters, Germany and France, came under EDP during the mid-2000s. Corrective actions under EDP were delayed, as MS referred to “exceptional circumstances” clauses in the SGP rules framework and MS failed to vote on stepping-up EDP actions against non-compliant MS. The Eurozone Economic Adjustment Programs (Slides 4-9) The combination of (i) largely absent market discipline and (ii) weaknesses in the implementation of EU-wide fiscal rules led to a build-up of substantial macroeconomic, fiscal and financial imbalances in EU MS over the 2000s. The financial crisis of 2008-2009 then provided a trigger for the unwinding of these imbalances, as markets re-evaluated the sustainability of MS budgets and economies were shook by the Great Recession in the US. In the process, several EU MS, including Ireland and Greece, lost market access to finance their budgets. In response, the IMF, the EC and the ECB stepped in and provided crisis countries with financial support through economic adjustment programs. In absence of an economic adjustment program, countries that lost access to private debt markets in 2009 would have had to resort to drastic and immediate consolidation of their finances and/or default on existing loan commitments. The Economic Adjustment Programs therefore provided program countries with temporary financing, allowing for a more gradual and orderly fiscal consolidation process. In return for regular disbursement tranches, program countries agreed on a set of policy reforms with the EC/ECB/IMF, covering fiscal, structural and financial issues. Public debt sustainability and the need for an outright restructuring of the public debt stock were also considered. These comprehensive economic reforms aimed at ensuring that program countries would re-emerge from crisis. To negotiate policy 10 Please see also “Playing by the Rules: Reforming Fiscal Governance in Europe”; Eyraud & Wu (2015); IMF Working Paper 15/67. 36 conditionality and monitor program implementation, program countries agreed to regular intensive “review” missions in their capitals by technical teams of the EC/ECB/IMF. Ireland is often seen as a successful example of such an adjustment program, exiting the program after 3 years. A property and banking crisis in the late 2000s led to government capital injections worth close to 30% of GDP by mid-2010. Both the direct fiscal impact of banking support as well as the ensuing macroeconomic recession sharply deteriorated public finances, leading the Irish government to lose market access. Its international partners responded by putting together a financial assistance program worth 85bn EUR by end-2010, conditional on a comprehensive banking stress test and financial reform as well as fiscal consolidation to ensure debt sustainability. Decisive implementation of the program allowed Ireland to re-access debt markets in 2012 and close the adjustment program after three years. In contrast, the Greek adjustment program is currently in its sixth year, with a further extension agreed in summer 2015. The source of the Greek crisis was an underlying deterioration of Greece’s fiscal position during the 2000s, partly hidden by insufficient fiscal reporting as well as a macroeconomic boom. As the full extent of these imbalances came to light in 2009/2010 and Greece lost its ability to place bonds on the market, the EC/ECB/IMF put together a first financial assistance program in mid-2010 worth 60% of GDP. In contrast to Ireland however, in the face of a sharp macroeconomic downturn and sluggish program implementation, the program had to be extended twice in 2012 and in 2015. Furthermore, a public debt restructuring was deemed necessary in 2012, which included a haircut on private creditors of 53.5%, a debt buy-back and significant rescheduling of interest and amortization payments.11 12 Despite differences in the speed of progress in Ireland and Greece, overall the programs have borne fruit. Both countries have executed a sharp improvement in their fiscal stance, there has been some progress on structural reforms and both countries have exhibited signs of a macroeconomic recovery. Nonetheless, unemployment remains stubbornly high. Public debt levels remain distressingly large, notably given spillovers from banking support operations. Following the election of the new Tsipras government in early-2015 and the negotiations for a 3rd Greek adjustment program, the ensuing uncertainty and temporary capital controls imposed in Greece are likely to undo some of the recent macroeconomic gains achieved in that country. Why has Ireland been able to execute its adjustment program so smoothly, while Greece is now entering its 3rd economic adjustment program? Beyond the fact that Ireland started with a banking crisis and Greece started with an underlying fiscal imbalance, the nature of the required fiscal effort has been significantly larger in Greece. Ireland was able to return to the market having (to date) improved its structural balance by 10pp (as % of potential GDP), 11 For further evaluation of the Irish and Greek programs, please see also “Ireland: Ex Post Evaluation of Exceptional Access under the 2010 Extended Arrangement”; IMF (2015); Country Report 15/20 as well as “Greece: Ex Post Evaluation of Exceptional Access under the 2010 Stand-By Arrangement”; IMF (2013); Country Report 13/156. 12 For further description of the Greek debt restructuring measures in February and December 2012, please see the relevant passages in “The Second Economic Adjustment Programme for Greece – March 2012”; European Commission (2012); pp. 47-48 as well as “The Second Economic Adjustment Programme for Greece – First Review December 2012”; European Commission (2012); pp. 52-55. 37 while Greece has improved its structural balance by 20pp and yet its finances remain in doubt. Equally, the nature of the Greek economy implies that a given budgetary consolidation effort is likely to have a larger impact on the economy – the so-called fiscal multiplier is estimated to be larger. The Greek economy is one of the most closed in Europe, while Ireland exports more than 100% of GDP. Because the Irish economy is therefore much less dependent on domestic demand, fiscal consolidation is likely to come at a proportionally smaller cost to economic activity. Equally, the Irish economy is characterized by a relatively high degree of flexibility- as suggested for example by its 13th place ranking in the World Bank Doing 2015 Business Report. As such, labor and capital dislocated through fiscal policy changes can re- allocate faster to new business areas.13 At the same time, Ireland has benefited from decisive implementation of the program, significantly lower policy uncertainty and a high degree of ownership over the reform program. While the financing envelope for Greece and policy conditionality were so far twice substantially renegotiated (in a 2nd program in 2012 and a 3rd program in 2015), Ireland executed its original economic adjustment program and then exited the program within the agreed timeframe. This relative stability of the macroeconomic and policy framework, in turn, allowed Ireland to run banking stress tests and recapitalize the banks upfront when its program started, while changing circumstances delayed the Greek banking sector recapitalization to two years after the start of the Greek program - dragging on uncertainty about the health of the domestic financial sector. The greater stability of Irish governments in this period also promoted determined program implementation. Overall, the smoother implementation of the Irish program may explain in part why the Irish documentation on policy conditionality agreed with the IMF has been substantially more concise than its Greek counterpart. Since the Crisis: Reform of the EU Fiscal Framework (Slide 10) As the EU is still resolving the current sovereign debt crisis, an effort is under way to strengthen the medium-term EU fiscal framework and to prevent a repetition of the current crisis. First, a range of measures strengthen ex-ante fiscal rules. These include a more explicit system of fines if MS remain under EDP. Recognizing the fact that macroeconomic booms and financial bubbles can hide underlying fiscal weaknesses, greater emphasis is put on achieving a sustainable structural balance as medium-term objective (MTO). An explicit Macroeconomic Imbalances Procedure (MIP) is added beside the existing EDP framework and fiscal monitoring is performed in the context of a more general review of national MS policies (the “European Semester”). Budgetary medium-term frameworks and projections need to be reviewed by an independent domestic watchdog agency (a fiscal council).14 13 For further discussion on the size of fiscal multipliers and how they depend on circumstances, see for example “A Simple Method to Compute Fiscal Multipliers”; Batini, Eyraud, and Weber (2014), IMF Working Paper WP 14/93 . 14 For further information, please see also “The EU’s economic governance explained”; European Commission (2014); Fact Sheet MEMO/14/2180 and “Economic Governance Review”; European Commission (2014); COM (2014) 905. 38 Second, the reform introduces a formal mechanism to deal with fiscal crisis in an MS. The European Stability Mechanism (ESM) is permanently established to provide financial support if MS lose market access. Third, recognizing the international nature of especially Eurozone banking, the new framework introduces a banking union reform. All Eurozone banks will follow regulation under the “single rule book”. The largest Eurozone banks will be regulated by a Single Supervisory Mechanism (SSM). In case bank resolution is needed, a Single Resolution Mechanism (SRM) will come into operation.15 Final Thoughts (Slide 11) To summarize, the recent experience in the EU suggests several lessons for the application of a rules- based fiscal framework in a federation of states: 1. Think beyond fiscal headlines: macroeconomic and financial imbalances can quickly spill over and drag down public finances. Risks to the budget need to be assessed taking a holistic perspective. Tackling macroeconomic and financial imbalances early on, before they build up, can help solidify public finances in the long-run. 2. Fiscal rules are only as good as the underlying statistics: investing in monitoring standards upfront helps ensure that fiscal rules can effectively enforce budget constraints. 3. Fiscal rules must “bite”: fiscal rules must be enforced. If violation becomes the norm, rules lose their social stigma and the fiscal rules lose meaning. A clear and credible menu of penalties must be in place to deal with deviations. Further lessons stem from the EU experience with economic adjustment programs: 1. Develop a credible upfront adjustment plan: the federation needs to jointly agree on a credible resolution plan once one of its members is in crisis (involving potentially short-term financial support, comprehensive measures to close budget holes and debt restructuring).16 This ensures that the economy does not “freeze up”, as businesses and consumers await the resolution of policy uncertainty. 2. Ensure that the government has ownership and take into account implementation capacity: swift and effective implementation of the adjustment program depends on ownership and capacity of the government in crisis. Without these two ingredients, programs are likely to get bogged down and extended – increasing the overall cost of the necessary adjustment. 3. Build a firewall to protect other members of the federation: crisis and adjustment programs will stoke fears about the budgetary situation of other members of the federation, potentially 15 Please see also: “Banking Union”; European Commission (2014); MEMO/14/294. 16 For consistency, this adjustment plan needs to properly account for the fiscal multiplier, the macroeconomic impact of fiscal consolidation. The IMF argues that these were systematically underestimated in the initial design of the Eurozone economic adjustment programs. Please see “Growth Forecast Errors and Fiscal Multipliers”; Blanchard & Leigh (2013); IMF Working Paper. 39 affecting their credit market access. Measures to contain the crisis, for example a federation-level financing agency that pledges to intervene if needed, can help stabilize the situation. References Batini, Eyraud, and Weber (2014); “A Simple Method to Compute Fiscal Multipliers”; IMF Working Paper WP 14/93. Blanchard & Leigh (2013); “Growth Forecast Errors and Fiscal Multipliers”; IMF Working Paper. Eyraud & Wu (2015); “Playing by the Rules: Reforming Fiscal Governance in Europe”; IMF Working Paper 15/67. European Commission (2012); “The Second Economic Adjustment Programme for Greece – March 2012”; pp. 47-48. European Commission (2012); “The Second Economic Adjustment Programme for Greece – First Review December 2012”; pp. 52-55. European Commission (2014); “The EU’s economic governance explained”; Fact Sheet MEMO/14/2180 European Commission (2014); “Economic Governance Review”; COM (2014) 905. European Commission (2014); “Banking Union”; MEMO/14/294. IMF (2013); “Greece: Ex Post Evaluation of Exceptional Access under the 2010 Stand-By Arrangement”; Country Report 13/156. IMF (2015); “Ireland: Ex Post Evaluation of Exceptional Access under the 2010 Extended Arrangement”; Country Report 15/20. World Bank (2014); “Doing Business: Going Beyond Efficiency”; World Bank Report. 40 Fiscal Rules and Debt Restructuring The European Union Experience Country Cases for Ireland and Greece Dr Christoph Ungerer & Dr Lili Liu Macroeconomics & Fiscal Management (MFM) Global Practice, Governance Global Practice, Equitable Growth, Finance and Institutions (EFI) Vice-Presidency 41 SECTION III - Addressing Subnational Indebtedness Problems – Distilling Lessons from Country Experiences “Priorities in the Short and Medium-term for China’s Subnational Government Debt: Good Practices and Options”. Prof. Roy Bahl, Regents Professor of Economics and Dean, Emeritus, Andrew Young School of Policy Studies, Georgia State University, USA “Debt Restructuring, Institutions, and Markets”. Dr. Lili Liu, World Bank, Governance Global Practice, EFI Vice-Presidency “Lessons for Borrowing Policy and Subnational Risks Management”. Dr. Sudarshan Gooptu and Dr. Karlis Smits, World Bank, Macroeconomics and Fiscal Management Global Practice, EFI Vice- Presidency “Fiscal Adjustment and Subnational Borrowing Regulations”. Prof. Jorge Martinez-Vazquez, Andrew Young School of Policy Studies, Georgia State University, USA 42 Priorities in the Short and Medium-term for China’s Subnational Government Debt: Good Practices and Options Roy Bahl17 INTRODUCTION Subnational government debt policy in China has been ad hoc and reactive to infrastructure needs and economic development incentives. Until 2014, provincial and local governments had little or no formal borrowing powers, but used a variety of “backdoor” approaches to finance infrastructure. However with the issuance of the new budget law in 2014, the government has taken the first steps toward rationalizing the use of debt finance by subnational governments. So far, however, the effort has been more focused on making provision for addressing the management of obligations incurred in the last decade than on putting in place the right set of institutions, incentives and controls for a sustainable framework. In this paper, I address the issue of how China gets from where it is now to a sustainable framework for subnational government borrowing. The next subjects to take up, I think, are the reasons why the present efforts should be focused on a transition program. A debt framework that can be sustainable in the long run will need to wait on structural and management reforms in four areas: (a) developing a strategy for controlling provincial and local government borrowing, (b) modernizing the approach to local government finance and to defining and enforcing hard budget constraints at the subnational government level, (c) establishing a more balanced approach to central and subnational government finances, and (d) identifying repayment options for subnational government debt. WHY A TRANSITION PROGRAM IS NEEDED The program of debt financing that China builds now will be a transition to a more permanent framework, and it will likely be in place for a number of years while important structural and management reforms are put in place to support effective subnational government borrowing. But infrastructure financing will not wait, hence the need for an interim regime. The transition program, which necessarily will emphasize direct controls, may be very different from the eventual permanent framework, which might rely more heavily on a combination of local discretion, market signals, incentives, and fiscal rules. Several factors make the transition approach necessary and advisable. Changing Economic Conditions The growth rate in the Chinese economy is projected to decline to about 6 percent by 2020 (World Bank and Development Research Center, 2014, pp81-126). Even though this is a healthy economic growth by world standards, it will not generate the fiscal surpluses that were seen in the past two decades. Expenditure management in this setting will present challenges. One particularly important issue is that an aging population will put pressure on pension and health outlays. The share of population aged 17 Regents Professor of Economics and Dean, Emeritus, Andrew Young School of Policy Studies, Georgia State University. This was a background paper prepared for the Summit. 43 60 years and over will rise quickly in coming decades, from around 12 percent in 2010 to almost 25 percent by 2030 and to more than 33 percent by 2050. This will not only place greater strain on family support networks but also will challenge social programs, pensions, and, in particular, health care (World Bank and Development Research Center, 2014, pp263-358). Expenditure claims such as this together with the expected lower level of revenues means that the nominal growth in some expenditure programs of central and subnational governments will need to be slowed, or new discretionary increases in revenues will need to be implemented. The implications are that increased pressure will be placed on local government revenues, and the residual available to finance new borrowing will be limited. Governments tend to be especially wary of economic policies that might discourage new investments and reduce economic growth in times when there are concerns about the performance of the economy and about job growth. If a new initiative is proposed, such as further separation of the roles of government and the private sector or reductions in tax incentives, there may be significant pushback or postponement of the new initiatives. This has already happened in the case of the State Council recommendations for halting local tax and subsidy programs for private enterprises. Improvements in the transparency of the budget, and measures to tighten the hard budget constraint, will be much more difficult in a time of concern about slow economic and job growth. Infrastructure Demands Slow economic growth or not, the demand for new infrastructure and for the renewal of existing infrastructure will be significant. Infrastructure demand will come from a number of fronts. Continued in-migration from rural areas will bring some pressure on capital finances. If the growth rate in real wages increases and the hoped-for growth in middle class consumption comes to pass, the composition of public services demanded could be pushed in the direction of better social services, a better natural environment, and better amenities. Changes in expenditure demand also could come from the business sector because of a shift away from manufacturing and toward a growth in the service sector. Pressures on the government expenditure budgets also will come from a need to address backlogs of unmet expenditure needs, debt and pension fund payments, and congestion and pollution problems. Other costs to be addressed include the maintenance costs of the increased public capital stock that has been built over the past decade and the higher cost associated with the urban sprawl that has come to characterize many Chinese metropolitan areas (World Bank and Development Research Center, 2014, pp 127-186). Some infrastructure investment might be delayed, but it will not be halted. A framework for borrowing by subnational governments will be required during the transition period. Cleaning up the Debt Backlog The estimated debt of subnational governments at the end of 2014 was about RMB 24 trillion (US$3.7 trillion), equivalent to about 38 percent of GDP. Presently, the government is focused on refinancing maturing debt to relieve the burden on current expenditure budgets. The program allows provincial and local governments to swap RMB 1 trillion in high interest and short maturity debt for lower interest and 44 longer maturity bonds. This could free up resources to support additional government borrowing or to address current expenditure needs. Oversight The transition plan will almost certainly emphasize direct controls on borrowing by subnational governments, as the State Council laid down in 2014. One should expect the central government to play a more direct role in determining which subnational governments can borrow, how much they can borrow, and for what purposes they can borrow. Until the role of the financing platforms is fully spelled out this will likely be regulated by the central government. The new budget law and the State Council Regulations of 2014 would seem to have called for the elimination of some elements of this practice, but the phase-out will take some time. The implementation of the full transition version of a debt framework will come well before the decisions about any new intergovernmental financing reforms, because the latter will involve major changes that will require time for design and implementation. Therefore, the transition framework will need to stay within the constraints of the present system and the new regulations given by the State Council. The central government will decide who can borrow and who cannot, and repayment of the debt will be from a “surplus” of unconditional grants and shared taxes over current expenditures and repayment of existing debt and pension/health fund obligations. Building a Debt Management Capability There is much to be done in putting together an administrative architecture that will manage the debt framework. The administration of a permanent framework that monitors the performance of subnational governments and the adherence to fiscal rules will be more complicated than a model that depends on direct controls. Development of the management capability to support a permanent framework is not presently in place and will take time to develop. At the subnational government level, debt management will also require a management infrastructure. The new budget reform will help because it mandates more transparency in all four local budgets. But matters will be complicated by the absence of a capital budget and less than complete transparency in the government fund budget. Efficient fiscal planning will be enhanced by the use of medium term fiscal planning networks. At the central (and/or provincial) level, a capability will need to be established to measure and compare the creditworthiness of subnational governments. Usually, “creditworthiness” has to do with assessing the ability of the local government to repay its debts fully and in a timely fashion. Governments are then ranked according to the ease with which they can meet this test, and are rated. The central government, or an independent rating agency, will need to do this assessment. If it is the central government, it will be necessary for it to set standards for creditworthiness, and to develop and update a data base for all subnational governments. 45 MARKETS, FISCAL RULES AND DIRECT CONTROLS Ultimately, China will need to decide on a balance between direct control of debt issuance (approvals on a case by case basis), fiscal rules that govern access to debt markets, and market discipline. However, even before it gets to the point of making that decision, the government must address several issues concerning the intergovernmental fiscal structure: (a) how to deal with the incompatibility of the present vertical imbalance in the system with the evaluation of debt repayment potential, (b) how to establish fiscal rules under the present intergovernmental fiscal system, (c) how to design and enforce incentives for fiscal discipline, and (d) how to reconcile the inherent conflict between eligibility to borrow and equalization. Most of these issues are not yet addressed, and likely will not be in the very near future. The enactment of a new borrowing regime will likely go forward under the present intergovernmental public financing system. Vertical Fiscal Balance The Chinese intergovernmental fiscal system is unbalanced. Most of the spending responsibility lies with the subnational governments, and therefore much of the demand for capital financing will lie with subnational governments. But almost all power to raise revenues lies with the central government. The exception is land lease revenues. In effect, the subnational governments have no capacity to repay debt other than that capacity given to them by the central government in the form or grants and shared taxes. Administrative Controls, Fiscal Rules and Incentives Any debt framework established for the transition period will almost certainly be characterized by direct administrative controls, and some fiscal rules, but how will these controls and rules be established? A first question to be raised is how debt will be rationed across and within provinces. If the criteria is revenue strength and the probability of meeting all debt service payments, then some lower income provincial and local governments will likely be excluded. But even this will be a difficult assessment to make because the only recurrent revenue source for subnational governments is intergovernmental transfers. At least for the interim period, the recourse will be to measure a fiscal surplus between “non - controllable” expenditure requirements and entitlement revenues from shared taxes. This will not be an easy calculation to make or to monitor. Fiscal rules could be used and fiscal indicators could be compared to assess ability to repay debt. For example, a “green light” community for new borrowing would be one whose debt burden is not high, who has fiscal resources that are high relative to expenditure needs, and whose economy is not vulnerable to cyclical swings and is not experiencing slow growth. Some measures of the quality of management of the local government might be added, based on their present practices. There are many questions to be answered about how to make the fiscal rules. For example, which sources of subnational government revenue should be used to make such a calculation? At present, the revenues available to subnational governments are shared tax entitlements and unconditional grants, earmarked grants, dedicated national taxes, and land leases. Of these, land leases are questionable as collateral for lending, and dedicated national taxes (e.g., the business tax) are vulnerable to central 46 government policy decisions. Earmarked grants are not available for repayment of general debt. This leaves only shared taxes and unconditional grants available as sources to cover debt repayment. A fiscal rule for the transition program might define a threshold for borrowing as the controllable portion of these resources as a percent of GDP. Rewards, Incentives and Penalties Debt policy could also include rewards, incentives and disincentives to induce more responsible fiscal behavior by subnational governments. The following might be considered, some for the transition period but others for the permanent program.  The fiscal rules could provide for increased borrowing in the event of increased revenue mobilization. If viewed in the traditional way, this would involve local governments increasing their tax rates or broadening their tax base to increase revenue mobilization. Subnational governments in China have no power to set the tax rates. However, Chinese local governments can expand the tax base, in two ways. First, they can use discretionary powers such as giving fiscal incentives to attract industry and thereby expand the tax base. Second, the tax bureaus in the provinces can follow the 2014 mandate of the State Council and collect full rather than partial liabilities and thereby increase total revenues.  Beyond the transition period, subnational governments could be given rate setting powers and could increase tax revenues as a local option. This would become part of the debt framework for the long run, and revenue effort relative to expenditure needs could become the basis of an important fiscal rule. There also could be disincentives built into the debt framework.  Borrowing by subnational governments could be restricted to government services, and could exclude any private sector activities. This might limit the demand for new borrowing.  Punitive measures could be put in place for subnational governments that defaulted on debt repayments. These might include an intercept of intergovernmental transfers, or replacement of lead provincial and local government officials. Fiscal discipline could be built into the regular evaluation of local government lenders.  Borrowing windows would be closed to subnational governments that did not meet the tests laid down in the fiscal rules.  Governments that did not meet certain tests of good fiscal management – timely approval of the budget, transparency of good budget reporting, separation of government from private sector activities – would not be given access to credit markets. Creditworthiness and Equalization Both the transition and the permanent framework will need to make provision for the problem of subnational governments with limited fiscal capacity. The creditworthiness approach will give borrowing power to subnational governments that can show stronger repayment potential and less risk of default. 47 But this approach often will rule out places with a weaker needs-resources balance, and will be inconsistent with equalization policy. How might this problem be addressed under the longer run framework? There are several ways.  Substitution of grant for loan financing,  Subsidize part of the repayment burden of the loan,  Build more equalization into the intergovernmental transfer system,  Reassign the responsibilities for some pro-poor expenditure programs to higher levels of government. HARD BUDGET CONSTRAINTS AND FISCAL MODERNIZATION China has recognized that its intergovernmental fiscal system has important weaknesses and that some reforms in the structure are necessary (Lou, 2013; Bahl, Goh and Qiao, 2014; Bahl, Goh and Qiao, forthcoming). Many of these needed reforms have to do with cleaning up the concept of a hard budget constraint, as applied in China. Three issues of particular importance are: (a) the right place for land leases in the financing structure, (b) the limited revenue discretion of local governments, and (c) the need for modernization of the subnational government fiscal structure and management. Land Leasing as a Revenue Source In theory, the sale of locally owned assets can be a source of financing for capital projects. Turning an asset that generates little current return into a part of the public infrastructure that generates a long term flow of benefits for the population can be good policy. China managed to use land leases to finance a significant amount of public infrastructure to accommodate a huge in migration of rural population. The problem with the sale of land leases to finance new infrastructure was not with the selling of a government-owned asset, but with the regime that went with this. It included (a) confiscation of land from farmers, (b) investment in some infrastructure that benefitted private sector enterprises, (c) the lack of transparency and uncertain liabilities, (d) ignoring the risk of sustainability if land values fell, (e) poor project selection, and (f) the accumulation of unnecessary amounts of land reserves. What to do in a new borrowing framework, either during a transition or in a sustainable borrowing framework? Land leases could be collateral for revenue bonds in financing a public project, or they could be part of the repayment guarantee for a general obligation bond (i.e., they could be part of the dedicated revenue stream). But, as has been learned in the past two years, the demand for land leases can be volatile and revenues can fluctuate widely. Local Government Revenue Discretion In a sustainable borrowing framework, as might be found in some industrial countries, subnational governments have the discretion to raise taxes. If expenditure needs rose to a level to challenge the possibility to repay loans, the tax rate could be increased. In this way the subnational government would be able to balance the budget and cover the fixed obligations. But under the present regime in China, local 48 governments do not have a formal discretion to increase taxes. They may, and have, used backdoor approaches to generate a flow of revenues to finance capital projects, e.g., extra budgetary revenues, UDIC financing platforms, tax collections that do not match tax liabilities, are examples. But all of these have proved to be non-sustainable. Even the intergovernmental transfers that support subnational government budgets have not proven to be sustainable. The income tax revenue shares have been reduced, and more recently the business tax and the agricultural tax have been abolished. Liberalizing the Intergovernmental Fiscal System In the fall of 2014, China’s top leaders presented a program for fiscal reform of the nation’s tax system, budgeting practices and center-local fiscal relations. The broad proposals are impressive, and a timetable has been set to modernize the fiscal system. The Minister of Finance has said that the major reform tasks concerning the fiscal and tax system will be completed by 2016, and a modern system will be fully developed by 2020. On August 31th of 2014, China's legislature adopted a revision to the Budget Law. This is the first time the Budget Law has been revised since it took effect in 1995. The old Budget Law was adopted when China was more strongly influenced by planned economy concepts, leaving a great deal of room for interpretation by governments. The amendments to the Budget Law are closely interrelated with China's fiscal reform. It took an unusually long time to revise it—seven years to draft a bill for the first reading in 2011 and four readings to get it passed. Subsequently in September of 2014, the State Council followed suit with directives on deepening the reform of budget and debt management. Among the key areas for reform are to improve legislation, clarify powers and responsibilities, restructure the tax system, stabilize tax burdens, produce transparent budgets, increase efficiency, and establish a modern fiscal system for both the central and local governments. In order to achieve these goals, the CCCPC proposes three major fiscal reform measures: improvement of the budget management system, improvement of the tax system, and establishment of a system of matching between administrative authority and expenditure responsibility. All are important to developing a sustainable framework for subnational government borrowing. Improve the budget management system. A completely standardized, open and transparent budget system will be adopted. The focus of budget review will be extended from fiscal balance and deficit size to expenditure budget analysis and fiscal policies. Most major expenditures that are presently indexed to increases in financial revenues or GDP will be delinked. This will allow a significant increase in budget flexibility. China also will establish a multi-year budget balance mechanism, a comprehensive government financial reporting system on an accrual basis, and a standardized and reasonable debt management and early warning risk mechanism for both central and local governments. The plan also includes integrating and regulating special transfer payments, and gradually cancelling funds that flow from local governments to the competitive sector. Improve the taxation system. Reforms in the tax structure will be implemented, and there will be a gradual increase in the proportion of direct taxes. The ongoing reform of the VAT will continue with the 49 folding-in of the business tax, and the tax rate structure will be simplified. The structure of consumption taxes will be upgraded and will be imposed on products that consume too much energy and cause serious pollution as well as some high-end consumer goods. Other elements of the tax reform plan include establishing an individual income tax system in which taxable income is defined in both comprehensive and categorized ways, passing new real estate tax legislation and implementing this in a timely manner, accelerating the resource tax reform, and changing the current environmental protection fee into an environmental tax. In line with the principles of unifying the tax system, promoting equality in tax burdens and fair competition in the private market sector, the management of preferential tax policies, especially regional preferential tax policies, will be rationalized. Legislation regarding the use of preferential tax treatments will be made clear and regulations will be issued. This will enable determination of which preferential policies can be continued and which cannot. In addition, the collection and management system for provincial and local government taxes will be upgraded. Comments on CCCPC decision by Minister Lou. 18 Finance Minister Lou Jiwei commented on these CCCPC decisions by pointing out that the budget management system reform, is the foundation and should be the first step taken in the reform. The reform of revenue distribution should be implemented after the substantial completion of tax system reform. Meanwhile, building a better system of matching between administrative authority and expenditure responsibility may require more time to form a consensus plan. While reform of the budget management system may make decisive progress in the short run, and reform of the taxation system may make significant progress in legislation and promotion in the short run, the reform of assignments of administrative authorities and expenditure responsibilities may only get as far as reaching a basic consensus. The major works and tasks of fiscal and taxation reforms will be basically achieved by 2016 and various reforms will be put in place, but a modern fiscal system may not be established before 2020. A NEW VERTICAL BALANCE? A transition framework for subnational government borrowing will be formed under the present financing arrangements, i.e., heavily decentralized expenditure assignments and heavily centralized revenue assignments. At present, most spending (over 80 percent) is made directly by subnational governments, but all revenue raising power (except land leasing) is at the central level. A sustainable debt framework for the longer run might benefit if China moved to a more balanced vertical fiscal structure, i.e., if there was a better match between the expenditure responsibilities for each level and their revenue raising powers. This discussion about revenue and expenditure assignments has begun, but the conclusions remain in the offing. Three components of a rebalancing might be considered: (a) assignment of more expenditure responsibility to the central government, (b) restructuring the revenue sharing system, and (c) giving more fiscal autonomy to subnational governments. Any of these three changes would have a significant effect on the borrowing framework. 18 As reported in Bahl, Goh and Qiao (2015), forthcoming. 50 Expenditure Assignment Expenditure assignments in China have not been changed since before the TSS reform in 1994. The present day Chinese economy is quite different from the one for which the present division of government expenditure responsibilities was cast. A thorough review of the division of expenditure responsibilities is long overdue. The end result of such a review could be a new budget law that will lay out the responsibilities for each sub function of government. Reform of expenditure assignments in China could usefully center on three areas. First, a more clear division between the role of government and the role of the private sector is needed. Second, the question of which level of government is best suited to finance and manage social insurance programs should be revisited. Third, there is need to review all of the functions of government in order to formalize and clarify the division of responsibility. On the first area, a reform direction that called for less direct involvement of government in the competitive sector would seem the right path. On the second area, both income distribution and labor mobility goals would seem to argue for more fiscal centralization. On the third, the maturing of the Chinese economy and the presence of spatial externalities suggests that there is need for more involvement of higher level governments in direct service delivery and/or coordination. Changes in expenditure assignments would change the borrowing needs of subnational governments, and could lead to changes in the fiscal rules or direct controls of debt issuance. And if finance did follow function, it also will require changes in the ability to repay debt. Revenue Sharing Changes in revenue assignment will be driven by the increased expenditure needs related to urbanization and by reassignments of expenditure responsibility. China faces the need to rebalance its system by shifting some functions to the central government level, and to provide for some sources of subnational government revenues to address the needs of provincial, city and county governments. The central government financing gap will be best handled by an increased retention rate of shared taxes. The subnational government revenue needs could be addressed by new local taxes. Both of these fiscal reforms would bring the Chinese intergovernmental fiscal system into better vertical balance. The case for subnational government taxation in China is a strong one. This initiative has always been off limits for Chinese intergovernmental fiscal reform, but perhaps the time has come. There are a number of benefits. Revenue mobilization would be enhanced because local governments have information advantages that give them a comparative advantage in the collection of certain types of taxes. Property and land taxes are a good example of this advantage. There also is a regional efficiency argument to be made. Substituting local taxes for some intergovernmental transfers would lead to a higher tax price in some local areas and force labor and capital owners to take this into account in making location decisions. China is struggling with urban sprawl. A system of property and land taxes could help rationalize land use patterns and provide incentives for more compact investments. Finally, there is an equalization argument, i.e., as local governments in higher income regions substitute own taxes for intergovernmental transfers, central funds will be freed up for distribution to lower income regions. 51 Empowering urban local governments to levy higher or lower tax rates would allow the fiscal system to preserve some of the innovation and the competitive strategy of local governments that has served China so well. The use of land lease sales in recent years and the use of extra budgetary charges in the 1990s and 2000s are examples of this resourcefulness. With taxing powers, local governments would have the wherewithal to expand the delivery of services that are in high demand or that would allow them to better capture their comparative advantage. It would also improve the creditworthiness for local governments by showing a stronger ability to service debt or to maintain public facilities. In general, local taxing powers would give local governments an instrument to use in shifting their competitive strategies from the back door to the front. Is Local Autonomy Necessary? If provincial and local governments issue debt under the current intergovernmental fiscal structure, repayment will be from central transfers to local governments. Nearly all sources of revenue are determined in amount by central government sharing rates and centrally-determined distribution schemes. Moreover, lenders should see some risk in this because central policies on intergovernmental transfers can change. See, for example, the recent decision to eliminate the major local government tax source by folding it into the value added tax. A major question in developing a debt framework for subnational government finance is whether ability to repay debt can be fully guaranteed by intergovernmental transfers, or requires some local discretion. SUBNATIONAL GOVERNMENT TAXING OPTIONS China has been slow to move away from its centralized system of public financing. But the problems with the centralized approach have become increasingly apparent. Subnational governments have sought (and found) backdoor approaches to do what they want, and sometimes this has led to undesirable outcomes, e.g., extrabudgetary funds and land lease collateral for unsustainable debt. When these backdoor approaches are withdrawn, subnational governments do not have autonomous local revenues to pledge to the servicing of debt. The question we raise here is whether there are feasible options for local government tax revenues. Our answer is that there are potentially revenue productive options that pass the tests of good international practice. Whether they pass the test for good governance in China is another issue. Property Taxation For three decades, property taxation has been discussed as a possibility for a major local government tax in China, but to date a broad-based annual property tax has not been introduced on a nationwide basis. This is not to say that real property is not taxed in China, but most of the existing taxes on real estate take place at the time of transfer and are better described as an ad hoc group of levies rather than a property tax system with clearly defined objectives. Together, these taxes on real property account for about 8 percent of national tax revenues, and 1.6 percent of GDP. The property tax has great potential as a revenue source in China. It is an opportune time to realize this potential. Most important, it could fill a revenue gap at the subnational government level, and could 52 be especially productive in larger urban areas. To get some idea of the revenue potential, note that industrial countries raise an average of more than 2 percent of GDP in property tax while low income countries raise about 0.6 percent of GDP on average. Advantages and Disadvantages. Because the ownership of real property wealth is usually concentrated in the highest income classes, the tax burden tends not to fall heavily on low income households. To the extent that higher taxes are paid on properties that tend to have better public services, the property tax might be seen as a benefit levy. Finally, since residential property taxes are capitalized into reduced property values, market decisions will be more rational because owners will pay a charge for the benefits of local services delivered, and speculators will face a real cost of holding their land off the market. The major problems with the property tax have to do with its difficult administration and its unpopularity with property owners. With respect to the former, good administration requires identification of all parcels and their ownership, valuation and revaluation, collection from individual owners, and it requires extensive recordkeeping and updating. All facets of the administration must be done well to realize the full revenue benefits from the tax. The political unpopularity, which seems disproportionate to the tax burden, has to do with the judgmental nature of the assessment, and the high visibility of the tax burden. Potential in China. China is one of the few countries in the world that does not levy an annual tax on residential property. In theory, China could generate considerable revenues with such an urban tax. An annual property tax that would yield the equivalent of one percent of GDP may be a feasible reform target in China. Revenue mobilization is of course an important objective, but the property tax also can be structured to encourage land use patterns that might lead to reductions in external costs. If the base of the tax is the assessed market value of the user right, for example, it could approximate an annual charge on location rents. The tax could be structured to provide an incentive for all users to develop and use land more intensively and could be made more consistent with an infill strategy for urban development. So would a tax on vacant property in the built up area that is imposed at a differentially higher rate. Again, this would not be a tax directly on the land, but on the market value of the user right. If the market value increased, so too would the tax. Liability would rest with the holder of the user right. With respect to property taxation, unused properties would be taxed at the same as occupied properties, at their value in highest use. Other property-related financing instruments might be useful in achieving the objectives of recapturing some of the value created by public investment, and increasing the density of cities. These might include CEPACs as used in Brazil, special assessments, betterment levies, valorization, and tax increment financing. Motor Vehicle Taxation Local government taxes on the ownership and use of motor vehicles could fit a Chinese strategy for coping with urbanization. The number of motor vehicles is growing faster than the population in urban 53 areas, their use imposes infrastructure costs on urban local governments, and they generate external costs that are for the most part uncompensated. Nearly 80 percent of the vehicle fleet is privately owned. The rapid growth in motor vehicles compared to that of the road network has been instrumental to the growing congestion levels and longer commutes. The result is a heavy external cost in terms of congestion and pollution. One study of Beijing has put the level of external costs at between 7.5 percent and 15 percent for all types of externalities (reported in World Bank and Development Research Center, 2014, pp187-262). Given this situation, it is understandable that Chinese policy has focused on controlling automobile use in large urban areas, and a good start has been made in this direction in a few cities. However, there is arguably even more to be gained by giving subnational governments additional control over motor vehicle taxation. The result could be to discourage the use of private cars, at the margin, while generating new revenues to defray some of the costs involved. Chinese cities could continue to use tax and charge policies either to increase the price of owning and using a motor vehicle, relative to using a public transport system, or to make a different housing location choice. This could be done by charging drivers for the full cost of using private vehicles, including environmental and social costs, through mechanisms such as higher registration fees, higher gasoline taxes, higher tolls and parking fees or various forms of congestion pricing. Depending on the price elasticity of demand, local governments could generate significant revenues from these charges and taxes. At present, Chinese subnational governments are not funded to any significant extent by taxes on motor vehicles. The registration tax on motor vehicles produces only a small revenue flow, and the tax on motor fuels is a central government levy. The experience with parking fees, toll and congestion charges as resource allocation measures is growing but has not yet become a major subnational government revenue source. The annual tax on motor vehicle registrations is based on engine capacity, as it is in most countries. At present, revenues from this tax are equivalent to only about 6 percent of that the present level of revenue received in VAT revenue sharing. The taxation of motor fuels is a more lucrative tax base, and it would more directly tax the use of motor vehicles. The question is whether such a tax could be locally administered in China’s larger cities. Ideally, the tax would be collected at the pump. Alternatively, it could be assessed at the factory gate as it is now, but the destination of shipments could be tracked. This would put in place a system whereby the rate of motor fuel tax could vary from one local government to another. Even at the current central government excise tax rates, a motor fuels levy would enable local governments to collect an amount about equivalent to one-third of VAT revenue sharing. Would it be fair to levy a higher tax on motor fuels in large cities? China’s gasoline prices are already above those in most developing countries, and above those in the US, but below the average in the European countries (World Bank and Development Research Center, 2014, pp439-535). On the other hand, Chinese gas prices are not high enough to cover the external costs of congestion and pollution. Would it be efficient to levy a higher rate of gasoline tax in China’s large metropolitan areas? At the margin, this would lead to a higher cost of living and would be a factor affecting migration decisions. 54 Higher gasoline prices would be consistent with other urban planning goals, such as reducing the congestion and pollution that come with automobile use, and raising the price of urban sprawl. In all of this, however, it is not so clear that the elasticities of substitution are high enough to make much of a difference unless there were very large rate increases. Local Retail Sales Taxes Local government sales taxes can be efficient instruments of urban government finance if collections can be shifted from an origin basis (place of manufacture or distribution) to a destination basis (place of consumption). In the United States, for example, a local government retail sales tax is collected at the point of consumption. Under an origin-based arrangement, such as China’s VAT, by contrast, collection is at the point of production and some of the tax burden may be borne by consumers who live in another jurisdiction. If administration on a destination basis could be made feasible, sales taxes on selected items of consumption could be revenue productive and would pass some of the tests of a good local tax. This might be done adequately for some items of consumption, for example, high end jewelry, durables and imported luxury goods. But for most consumer goods, a retail sales tax would encourage tax avoidance by providing an incentive to shift the point of consumption to informal traders that are not easily policed by the tax authorities. The problems with collection from informal sector retailers is also the reason why the revenue productive excises cannot be converted to retail sales taxes with higher rates imposed on consumption in cities. Otherwise, particularly beer, liquor and cigarettes would be attractive revenue bases. There is another option. All of the excise revenue could continue to be collected as at present, but the revenues could be assigned to cities on some formula basis. This would be a transfer of revenue, but would not give the local governments any power to increase or reduce rates, or to differentiate their rate from that in other urban areas. In effect, it would be another tax sharing transfer. Piggyback Taxes Some of the administrative problems with local taxation can be avoided by adopting a “piggybacking” approach, i.e., to allow the local government to select a tax rate to be imposed on a central government tax base. Piggybacking is already done in China with the urban construction and maintenance tax and the education surtax. In these two cases, the base is the aggregate tax collections from VAT, excise taxes and the business tax, and the revenues from this surtax are allocated fully to the local government. The sur- rate, however is set by the central government. The piggyback tax proposed here might be viewed as a benefit charge, i.e., a payment for using city services. Under such a reform, the existing UMCT and education surcharge would be replaced with an “urban service tax”. The local government then could be given the option of increasing the piggyback tax rate above its present level. To demonstrate, let us suppose that the tax rate on UMCT was increased from 9 percent to 12 percent. The additional revenue generation would be large enough to recoup an amount equivalent to about 44 percent of the shared VAT revenues. Similarly, a piggyback personal income tax could generate enough 55 new revenue to recover an amount equivalent to 9 percent of VAT revenue sharing. There are good arguments to use individual income tax revenue sharing, including the lower likelihood that the burden will be exported. But the already high tax rate on labor is a major drawback. Unless the social security contribution rate can be lowered, this is probably not a viable option. CONCLUDING THOUGHTS China could benefit significantly from an overhaul of its intergovernmental financing system. One benefit is that such a reform could help rationalize a sustainable framework for subnational government debt. The deliberations behind changes in provincial and local government expenditure and financing structure will take time. But the development of the national infrastructure will not wait until all of the needed structural changes to the intergovernmental fiscal system are put in place. For this reason, China needs (Beijing: China Financial and Economic Publishing House) a transition plan for capital financing. The transition framework for borrowing needs to be recognized as just that – a bridge from the present system to a sustainable long term framework. Even though the transition debt policies could be in place for a number of years, the Chinese Government would benefit if it had its long term objectives for subnational government borrowing in place when it develops the transition plan. References Asian Development Bank (2014) Money Matters: Local Government Finance in the People’s Republic of China (Manila: ADB). Bahl, Roy, Chor-Ching Goh and Baoyun Qiao (2014). Reforming the Public Finance Sysem to Fit a More Urbanized China (Beijing: China Financial and Economic Publishing House). Bahl, Roy, Chor-Ching Goh and Baoyun Qiao (forthcoming) ”Options for Reforming Shandong’s Fiscal System” (Beijing: China Financial and Economic Publishing House). Bird, Richard. (2013). “The VAT as a Local Business Tax” Tax Notes International, September 16, page 1-7. Liu, Lili and Baoyun Qiao (2013). “Restructuring of Legacy Debt for Financing Rural Schools in China” in Until Debt Do Us Part: Subnational Debt, Insolvency, and Markets edited by Octaviano Canuto and Lili Liu (Washington DC: The World Bank), pp. 81-108. Lou, Jiwei. (2013). Rethinking of intergovernmental fiscal relations in China. (Beijing: China Finance and Economics Press). Man, Joyce (2013). “Evaluation of Property Tax Reform and Experiments in China” Unpublished paper prepared for the Lincoln Institute of Land Policy (Cambridge, Mass), August 23, 2013. Smolka, Martim (2013). “Implementing Value Capture in Latin America: Policies and To ols for Urban Development (Cambridge, Mass: Lincoln Institute of Land Policy). World Bank and Development Research Center (2014). Urban China: Toward Efficient, Inclusive and Sustainable Urbanization (Washington DC: The World Bank). 56 Debt Restructuring, Institutions and Markets Lili Liu Debt Restructuring, Institutions, and Markets Dr. Lili Liu Global Lead, Decentralization and Intergovernmental Finance Governance Global Practice Equitable Growth, Finance and Institutions Vice-Presidency The World Bank October 22, 2015 57 Lessons for Borrowing Policy and Subnational Risks Management Sudarshan Gooptu and Karlis Smits19 China is not alone in its build-up of subnational liabilities to meet its development needs. What is more noteworthy has been the recent high rate of increase in this indebtedness of Chinese Provinces. Cross- country lessons suggest that a Subnational Debt Workout with the Center is just a start of a debt resolution process. The policy and institutional reform measures that accompany it, which restructures the subnational entity itself, is what will bring debt and fiscal sustainability, equitable growth, and economic strength to the region, and the country, over the medium term. Fiscal Rules have played in important role to try to maintain fiscal discipline and avoiding debt problems in future. Linkages between Medium-term budgets (MTFF), Debt and Fiscal Sustainability (DSA/FSA), and Medium-term Debt or Borrowing Strategy (MTDS) are important as well. Subnational Debt is only one form of Contingent Liability that can pose fiscal risks for the Government. There are others as well – Implicit and Explicit, Direct and Indirect. Thus, a consolidated view of these fiscal risks will provide a way for the government to minimize fiscal surprises in future. Subnational Debt is inevitable While urbanization and decentralization typically increases spending responsibilities of subnational governments (e.g. in provinces, and municipalities), the opportunities and constraints facing subnational governments when making their fiscal and debt decisions differ somewhat from those facing the Central government. Specific institutions and legislation shape policies at subnational level, most notably the revenue-sharing system, the intergovernmental transfers, and the implicit national guarantees that are accorded on local government debts. For instance, subnational governments typically give up tax bases as part of federal arrangements (e.g., they cannot impose interstate import/export duties). They often rely on transfers from the national government that are earmarked, must be allocated to specific spending programs, and cannot be used arbitrarily for budgetary consolidation and adjustment. Or, on the contrary, there may not be well-institutionalized transfers that depend on relative bargaining power between national and subnational entities. As a result, local governments typically face borrowing costs that underestimate the true credit risk associated with economic and fiscal fundamentals because investors price-in implicit guarantees provided by the national government. The 2008 global crisis brought with it an increase in subnational deficits and debt build ups in OECD countries. This was a result of the combined impact of decreased subnational tax revenues and increased demand for social services (which are typically provided in most OECD countries by subnational governments). Subnational Government debt in OECD countries grew from an average of 8% of GDP in 2007 to 10% of GDP in 2010. There is a lot of diversity across countries with subnational debt accounting for 1% of total public debt in Greece to 53% of the total in Canada. From 2010 onwards, when OECD national governments started implementing fiscal consolidation measures and withdrawals of the special 19 The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations or to members of its Board of Executive Directors or the countries they represent. 58 fiscal stimulus measures that were taken by Central Governments in the aftermath of the 2008 crisis, this tended to reinforce this trend in their subnational levels (Vammalle, 2013). Aside from its own source revenues (e.g. from subnational taxes and fees), typically a local government has several sources of financing it spending needs (and associated budget deficits) in any given year (Figure 1). It includes transfers from the Central Government (often guided by intergovernmental fiscal legislations, or on-lending of foreign loans taken on by the Central Government), special development funds (for nationally important and strategic priority public investment projects), loans from state-banks or private banks, and from issuance of debt securities in domestic or external markets. Figure 1. There can be many sources of direct financing for local governments In China, local government expenditure growth has tapered sharply in the first two quarters of 2015. Expenditure growth is decelerating because local government fiscal space is constrained, as revenues from land-lease transactions slow and reforms in the budget management system reduce quasi-fiscal activities. Issuance of new debt to finance new projects is moderating somewhat as the authorities introduce guidelines on strengthening management of local government debt. Corporate bonds issued by local government financing vehicles (LGFVs) to fund municipal investments slowed sharply in the first quarter of 2015. LGFVs issued only 194 billion RMB, a 45 percent year on year decline (figure 2). However, in recent months activity has accelerated as some of constraints to issue corporate bonds by LGFVs were relaxed—in May 2015 alone 121 billion RMB worth of such bonds were issued. At the same time, the authorities continue to make progress in decoupling the use of local government debt and provision of public infrastructure, such that growth of fixed asset investment in infrastructure is no longer strongly correlated with an increase in the net debt position of LGFVs. 59 Figure 2 Local government capital expenditures decline sharply in Q1 and Q2 as fiscal revenues from land-lease transactions slow sharply and quasi-fiscal activities are brought under control Growth of land-lease revenues and Local Growth of net position of LGFV corporate bonds, Government Fund Budget expenditures, percent billions of RMB (year on year) 80% 60% 40% 20% 0% Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 -20% 2013 2014 2015 -40% -60% Land lease revenues Local level Fund expenditures On 29 August, China’s National People’s Congress released the State Council’s update of debt statistics for sub-national governments. The update shows that sub-national debt at the end of 2014 rose by more than one third to RMB24 trillion ($3.7 trillion), the equivalent of 38% of China’s 2014 GDP, from RMB17.9 trillion as of June 2013. Higher levels of both direct and indirect debt suggest that RLG indebtedness has increased on account of new borrowing since June 2013. Direct RLG debt obligations rose 41%, while indirect and contingent liabilities rose 23%. At the same time, the National People’s Congress capped the total amount of RLG direct debt at RMB16 trillion, which equals total direct debt of RMB15.4 trillion at the end of 2014 and RMB0.6 trillion of new bond issues in 2015. 60 Figure 3 Total outstanding sub-national debt remains high but efforts to refinance maturing debt at lower maturities is underway Total amount of outstanding sub-national level debt Total amount on debt swap ceiling and and contingent liabilities cumulative amount of debt-swap bonds issued 30000 25000 20000 Billion RMB 15000 10000 15400 16000 5000 10886 0 End June 2013 End 2014 Est 2015 Guarantees and other contingent liabilities With direct payment responsibility Source: World Bank Staff estimates from MoF data. China’s Subnational Debt Resolution Process is just getting underway These recent revisions to the Budget Law are a big step in advancing China’s fiscal reform efforts toward a modern fiscal system. They lay the foundations for the overhaul of the fiscal system and should improve the quality of budget management and local governments’ debt sustainability over the medium term. The measures also facilitate efforts by the central government to align local governments' objectives more directly with the aim of public service delivery. Ultimately, these reforms are expected to channel Local Government Financing Vehicle (LGFV) debt to a more transparent bond market and encourage the use by SN governments of market-based instruments to meet their future financing needs. The recent change in the Budget Law is also an important first step towards resolution of China’s subnational indebtedness situation. On August 31 2014, the National People’s Congress adopted revisions to the budget law to enhance transparency of the budget, improve the budget control mechanisms, strengthen accountability for local government debt management, and improve the intergovernmental transfer system. This is the first time the budget law has been revised since it took effect in 1995. The old budget law was adopted when the country was still strongly influenced by planned economy concepts and at an early stage of applying budget management practices. It had only a very general definition of what the government budget covers, leaving ample room for interpretation by local governments. China's government budgets only included the general budget, largely made up of tax revenue and spending on public services and government operations. The revenue gained through transactions in use rights of state land, which account for a large part of local government revenues, and finances of state-owned enterprises were not budgeted or supervised by provincial people's congresses. 61 The revised law, however, defines the government budget in four parts: the general budget, the budget for government-managed funds, the budget for state-owned assets, and the budget for social insurance funds. All four are now under formal control of legislatures. With a view towards achieving greater budget transparency, the revised law introduces detailed provisions to ensure the public's access to government budget information. The old law did not have any provisions about public supervision. The revised law requires government financial departments to publish their budgets and final account reports within 20 days after they are adopted by the legislature. It also gives the public access to information about local government debts, purchases, budgets and audits. The new law also clearly regulates how the legislature to examine the government budget report and what to examine. It specifically asks the legislature to examine major expense items and big investments as well as to inspect the development and efficiency of such projects. Previously, the Budget Law severely restricted local governments from issuing bonds on their own credit profile; hence, many local governments relied on off-balance sheet entities to raise funds, mostly to fund infrastructure. This practice severely undermined prudent debt management. The revision allows for bond sales by provincial-level governments but places them under strict conditions. It not only restricts the amount of bonds but also regulates how they can be issued and how the proceeds can be used by subnational governments. Under the new version, provincial governments are now allowed to issue bonds within a quota set by the State Council, China's cabinet, and approved by the National People’s Congress or its standing committee. Money raised by the bonds can only be used for public services, and not for government operations. The debts must be reflected in the provincial budget and supervised by provincial people's congresses. The central government will assess risks inherent in this local debt. As a result, the revised law has helped clarify several important principles on how to manage local debts, setting rules and dividing liabilities. In addition, policy measures have been introduced to address the most immediate funding pressures of China’s subnational governments. The authorities have indicated that a reasonable grace period will be set for current projects to prevent the fiscal and financial risks that could be associated with a sudden cut in financing for the highly indebted subnational governments. The MOF’s October 2015 report to the National People’s Congress (NPC) (Ministry of Finance, China, 2015) indicates that the Central Government will allow the ongoing SN projects that are already under construction to be refinanced through commercial bank loans (most of which would be expected to from state-owned banks).20 Efforts to refinance existing debt at lower maturities is underway. The original plan announced in early March 2015 stipulated that provinces and municipalities would be primarily responsible for issuing and repaying the new debt using market mechanisms, such as bond auctions. In May authorities clarified details of the swap indicating that local governments can instead also negotiate directly with creditors of those swappable debts. Interest rates for these new bonds will also negotiated by debtors and creditors, 20 This measure supersedes the earlier P.R. China’s State Council issued Debt Management G uidelines (of October 2014) which made such bank refinancing in doubt. 62 but should not be lower than the prevailing treasury yields of the same tenor or exceed 30 percent of the treasury yields of the same tenor. Local governments have refinanced debt in amount exceeding RMB 1800 billion in a short period between mid-May and August 30, 2015. Most of these bonds were issued using bond auctions with average coupon rates ranging from 2.89 percent for a three-year bond to 3.56 percent for a seven year bond. It is planned that these new bonds will be allowed to be used as collateral with the central bank for repo and other liquidity operations. Box 1. Local government debt swap in China, 2015. On March 14, the NPC approved the 2015 budget, which covers important measures for local government debt. First, the budget set a RMB 600 billion bond issuance limit for 2015, an increase of 50 percent from 2014. This limit is split between RMB 500 billion for general obligation bonds and RMB 100 billion for special-purpose bonds tied to specific projects. Second, the MOF will permit regional and local governments to swap RMB 1 trillion of high-interest and short-maturity debt raised through third parties into low-interest and longer-dated bonds. According to the MOF, the debt swap represents “a change of form” and “not a rise of outstanding debt”. The initial RMB 1 trillion debt-for-bond swap program is designed from a 2013 local government debt audit by the National Audit Office. The program does not take into account debt accumulated after June 30, 2013. All 31 provinces and five centrally planned cities will be allowed to issue new bonds worth up to 53.8 percent of their debt falling due in 2015 as of June 30, 2013. Use of this latter date implies that the MOF has not finalized reclassifying local debt with the latest information that local governments submitted by January 5, 2015, possibly on concerns that these January figures overstate the true outstanding debt. The initial debt-for-bonds swap covers just a portion of all outstanding debt and contingent liabilities of local governments estimated to be around 18 trillion RMB at the end of 2013. It also covers a small share of potential payment responsibilities in 2015 if some of guarantees or contingent liabilities are called upon. To address these concerns, in early June authorities announced the launch of another RMB 1 trillion debt-for- bond swap for local governments in May, and expanded the program by an additional RMB 1.2 trillion in August, 2015. This will provide an additional buffer to local governments in executing an orderly rollover of maturing debt. Source (Ministry of Finance, China, 2015) The RMB 3.2 trillion debt-for-bond swap program (See box 1) addresses one of the key short-term vulnerabilities—namely, that without an orderly rollover of its outstanding debt, local governments’ own resources are insufficient to fully service maturing debt liabilities. The debt-for-bonds swap covers just a portion of all outstanding debt and contingent liabilities of local governments, which were estimated at around RMB 18 trillion at end-2013. However, the program addresses the most immediate concerns with respect to maturing debt in 2015 and clarifies recent regulations. The program could also cut local governments’ annual interest expenditures by about RMB 100 billion, implying a de-facto reduction in the average interest rate on outstanding LGFV bonds of about 500 basis points. Similarly, using local government bonds instead of corporate bonds to fund new spending would result in lower interest expenditures. For example, the average interest rate spread between seven-year bonds issued by LGFVs (AA-) and local governments averaged about 320 basis points in the first five months of 2015. 63 Fiscal Rules and SN Debt Control Mechanisms Subnational governments are often subjected to formal arrangements enticing them to pursue fiscal discipline and budget transparency. In OECD countries, SNGs are subject to strict fiscal rules and so their increase in indebtedness has not been as high as compared to their Central Government’s deficits and debt during the aftermath of the crisis. In many countries SNGs can only borrow to finance public investment and/or face tight ceilings on their debt and debt service (e.g. Maximum level of annuities as a share of operating revenues.). Since the institutional specificities of subnational governments create fiscal risks for them as well as for the Central government, there is a wide spectrum of subnational fiscal rules and ceilings on spending, deficit, borrowing and debt stock. Box 2 presents some international evidence concerning subnational debt limits, which have played a key role in achieving subnational debt sustainability. Box 2. International experience on subnational debt limits. Brazil. The Fiscal Responsibility Law sets a limit of 200 percent on the debt-to-net current revenue ratio (current revenue net of certain national government transfers), so there is a legal limitation on the amount (not the use) of new debt flows. It also imposes a limit of 11.5 percent on the debt service-to-net current revenue. Debt restructuring agreements between federal and state governments established a comprehensive list of fiscal targets. Colombia has extensive legislation on government debt which established a traffic-light system to regulate subnational borrowing. States in the red-light zone, with the debt-to-current revenue ratio above 80 percent and the interests-to-operational savings ratio exceeding 40 percent, are prohibited to borrow. India imposes a ceiling on budget deficit-to-GSDP ratio of 3 percent for all states and requests them to gradually close the current deficit. In addition, those whose debt service exceeds 20 percent of revenue are deemed to be at ‘debt stress’ and the national government closely monitors a ll new borrowings. Poland sets a limit of 60 percent on the debt-to-revenue ratio, restricting the amount of indebtedness, and a limit of 15 percent in the debt service-to-revenue ratio. As soon as the local government debt exceeds 55 percent of revenue, however, the debt service must be reduced below 12 percent of revenue. United Kingdom. After the September 2014 referendum in Scotland where its citizens chose to remain part of the United Kingdom, thereby, continuing to be part of the UK single currency (Pound Sterling) and single market, have a unified pensions system, and collective security and being part of the sixth largest economy in the world. The UK Government has subsequently just revised its Scotland Act 2012 on the intergovernmental fiscal devolution arrangements and outlines the executive borrowing and income tax powers already devolved to UK’s “family of nations”. This new Scotland Bill, which will come into force in 2015 and 2016 respectively, will ensure that, for the first time, over 50 per cent of the money spent by the Scottish Parliament will be funded by revenues raised in Scotland. This aims to deliver greater accountability, allowing more decisions affecting Scotland to be made in Scotland. As a result, the Scottish Parliament will have one of the most extensive arrays of tax and spending powers of any devolved parliament in the developed world. United States. The US states mainly follow golden rules constraining the operating balance to be zero, thus debt must be used to finance capital expenditures. Rules governing subnational borrowing vary across states and depend on the type of debt issued, the revenues used as collateral, and the type of government entity issuing debt. Markets play a key role in fiscal surveillance. Source: Juan Pradelli (2014), “Subnational Fiscal Sustainability and Debt ceilings” DMF Knowledge Note, World Bank. 64 When deciding on choice of fiscal rule to encourage prudent subnational government borrowing to finance its public investment needs, there is no one right answer. It all depends on the risk tolerance levels of the policy makers and the sources of revenues for the subnational government (tax and non-tax). Most countries have fiscal balance rules and debt rules but implementing expenditure and revenue rules is challenging for all, more so for emerging economies and low income countries. Expenditure rules are more prevalent in advanced and emerging resource-rich countries (Figure 4). Figure 4. Most countries have either debt rules or balanced budget rules. Source: Schaechter, Andrea, Tidiane Kinda, Nina Budina and Anke Weber (2012). The following is an example of four alternative fiscal expenditure paths that a resource-rich subnational government examined to meet a policy determined fixed composition of public investments over the medium term. Similar approaches that have been adopted by other countries are: Spend-as-you-go (SAYG): This approach presents the least strategic policy stance. In this case, the entire flow of resource revenues are spent on additional public investment projects as they become available. Permanent Income Hypothesis (PIH): This approach denotes a sustainable spending path that allocates the spending evenly over the time. Public investments are scaled up by an amount equal to the permanent income annuity implied by the present value of resource wealth. Bird-in-Hand (BIH): This approach represents the most stringent spending stream. All revenues from the oil sector are accumulated in a sovereign wealth fund that invests abroad. Only the interest earnings from this fund are used to scale up public investments. Moderate Frontloading (MF): This approach presents a case where public investments are scaled up rapidly in the early phase of the simulation period. In principle, neither the flow nor the stock 65 of additional spending stream is linked to the resource wealth under this approach. However, in the long-term, the amount that is used for additional public investments converges to the Permanent Income annuity to ensure fiscal sustainability. The spend-as-you-go (SAYG) approach does not lead to any savings. So, transfers to the budget due to revenue surpluses diminish over time, e.g. following a resource revenue depletion. Under the PIH, the government transfers a fixed amount of funds to the fiscal budget annually (black line in panel a). In the short term, this may indicate subnational borrowing (the first red shaded area) as resource revenues are considered relatively low at this stage. In the medium term, the resource revenues picks up and reaches a peak (blue curve). The difference between revenues and transfers in this hypothetical situation are saved in a contingency/stabilization fund (green shaded area). Finally, as the revenues gradually die out, interest earnings on the contingency/stabilization fund assets are used to supplement the transfers to the budget (second red shaded area). Under the BIH, transfers to the fiscal budget are scaled up over time as resource revenues are saved in the sovereign wealth fund and interest earnings on wealth fund assets increase (black curve line in panel b). As long as resource revenues exceed the transfers for the earmarked public projects, therefore, reserves continue to build up. Later in the projection horizon, accumulation comes to a halt and BIH annuity reaches to a plateau. Finally, the “big push” under the moderate frontloading approach leads to investments that are financed by borrowings in the short-term (first red-shaded area). In the medium term resource revenues exceed the spending, however the difference is smaller than under the PIH and BIH. Moreover, the spending converges to the PIH annuity in the long-term; however, they are always above the PIH. Therefore, contingency/stabilization fund savings would be a lot smaller than the PIH and BIH levels21. Whether there are fiscal rules on subnational governments or not, cross country evidence suggest that having a Medium-term Fiscal framework (MTFF) and regularly undertaking Debt Sustainability Analysis (DSA) at the subnational level, even before sovereign guarantees are issued and funds borrowed from any source, demonstrates fiscal prudence and sound debt management practices. These are useful analytical frameworks and tools for assessing the envelope of revenues and borrowings that subnational governments need to undertake urban infrastructure investment and provide decentralized social services such as education and health. It also allows the Government to assess how much fiscal adjustment will be needs in order to return a subnational government’s debt situation into a sustainable path in case of debt servicing difficulties. It also enables the setting up of an institutional structure in government where the different relevant government departments (at Central and subnational levels) formally collaborate and exchange information regularly to undertake these DSAs and better track emerging subnational debt-related vulnerabilities on a continuous basis. 21 See Scheachter, Andrea, Tidiane Kinda, Nina Budina and Anke Weber (2012) for a more detailed description of and country examples of various fiscal rules. 66 The impact of fiscal rules has been mixed as well, depending on the dimensions the government wants to focus on. The OECD has recently designed a Subnational Fiscal Rules indicator that examines the effectiveness of a country’s subnational fiscal rules on the basis of success in meeting four preferred attributes, namely, (i) restraining the size of the public sector, (ii) supporting allocative efficiency, (iii) ensuring debt sustainability, and (iv) coping with shocks. OECD Decentralization data for 2011 for selected countries shows that fiscal rules have been more effective in supporting allocative efficiency of public spending by subnational governments. It has been less so in restraining the size of the public sector and excludes operations (and growth) of quasi-fiscal activities at local levels that resulted instead. (OECD, 2015) Ensuring budgetary transparency and discipline Cross-country experiences suggest that any sovereign guarantees (whether issued at central or subnational levels) should be identified, registered and disclosed in financial statements. They should be quantified and included in the state budget as an expense item. The Central Govt. may charge the beneficiary a fee for the guarantee. In some cases (e.g. Sweden), a notional Contingency Fund is created to provide transparency of expected pay outs when contingent liabilities are triggered. General Contingent Liability Policy can be incorporated in the legislation. Such a law (e.g. A Public Debt Law or a Fiscal Responsibility Act) should at least impose an annual limit on guarantees that can be issued, and specific limits on overall indebtedness levels of the subnational entity. Main building blocks of a Fiscal Risks Management Framework include creating general policy for government exposure to Sovereign Guarantees and other Contingent Liabilities (CLs), ensuring budgetary transparency and discipline continuously applying financial risk management principles and examining cost-risk tradeoffs of different financing alternatives. To this end, how one measures debt and deficits for policy decision-making purposes matters. This is intrinsically linked to the definition of “government” that is used to report liability positions in the budget. For example, the U.S. federal government’s deficit in fiscal year 2010/11 was either 8% of GDP or 14% of GDP depending in whether it was estimated on a cash or accrual basis – the latter accounting for the impact of explicit contingent liabilities of the federal government while some researchers, namely Kotlikoff and Burns say that the “true” US budget deficit for that year was 39% of GDP. (Irwin, 2015) If the definition of subnational government in deficit and debt targets is too narrow, it may encourage shifting of spending through entities outside the defined perimeter of Government (and result in operations involving off-budget assets and liabilities). Wider definitions in turn may lead to errors in measurement of on-balance assets and liabilities. As a result, these will influence any assessments of a government’s spending and tax policies and resulting fiscal sustainability over time (Irwin, 2015). It is therefore prudent to be cognizant of the fact that estimates of subnational indebtedness are often underestimates which do not account for debts that emerge as contingent liabilities that result from the build-up of “hidden deficits” over time though the use of off-budget financing vehicles and/or inadequately accounting for the “true” government obligations under private public partnership arrangements in public investment projects, especially in infrastructure. Mispricing of collateral assets in subnational bond issuances (e.g. land) and implicit liabilities that emerge from perceptions of future expected government bailouts due to systemic considerations (e.g. “too-big-to fail” financial institutions 67 that may be owned by the subnational government themselves). The recent post-2008 global crisis experience of the European Union and continued discussions within it on the need for a fiscal and banking union to complement its monetary union in the EU area vividly highlights how debt resolution and financial sector health are intricately linked as well. Figure 5. Europe: NPLs after the global financial crisis (Percent of sample assets) In designing a framework for contingent liabilities management, any overall policy (for issuance of sovereign guarantees, for instance) should clearly establish conditions for risk-sharing and overall policy of risk allocation between the central, subnational and private sector entities involved in the transaction. What are the government’s reasons and priorities for providing such guarantees/contingent support? Why it sees the need to issue such contingent liabilities rather than providing direct loans or subsidies to these public sector/subnational entities? What types of risks willing will it cover and what can be shifted to the private sector? Best practices in monitoring subnational fiscal risks from contingent liabilities suggest that in publications of budget documents, explicit contingent liabilities of the government are identified, registered, disclosed; also quantified (as expected cost/ maximum probable loss). Depending on institutional realities and risk tolerance of the government, the fiscal costs could be directly budgeted and reserve funds created on the basis of established financial risk management principles. Ideally, quantification of cost and risk of subnational loans and other contingent liabilities of the government through different techniques can be undertaken subject to availability of reliable and consistent data so as to estimate how contingent payouts will affect country’s fiscal situation. Subnational Debt is only one source of fiscal risks to Government While the focus of this paper has so far been on Subnational Debt, it is important to note that a comprehensive picture of the true fiscal risks that the Government is exposed will need to look at its contingent liabilities beyond direct borrowing by subnational entities. This includes an assessment of the sources of fiscal risks emanating from the government’s direct and contingent liabilities, and whether they are explicit or implicit liabilities. The “Fiscal Risk Matrix” is one such analytical tool developed at the World 68 Bank that allows for such an assessment (Polackova-Brixi, 2002). A consolidated view of these fiscal risks will provide a way for the government to minimize fiscal surprises in future. Table 3. The “Fiscal Risk Matrix” – A way to catalogue the sources of fiscal risks to Government Source: Contingent Liabilities – a threat to fiscal stability, PREM Notes, n.9, November 1998. Note: These liabilities refer to fiscal authorities in central government, not the central bank. On the asset side, risks erode the value of accumulated assets and reduce the inflows. This also needs to be considered when managing fiscal risks in a Government. This calls for regular monitoring of contingent liabilities and conducting scenario analyses and stress tests on the government’s debt and fiscal sustainability situation over time. One approach to monitoring CLs is using an asset-liability management (ALM) approach but this is more related to financial management of contingent liabilities. One must continuously keep an eye on debt and fiscal sustainability continuously. Good experiences in LAC: Brazil, Chile, Colombia, Mexico, and Peru. Monitoring of CLs is done here in the context of the Fiscal Responsibility Laws, highlighting fiscal impact of CLs and not just financial impacts. • Colombia – Transparency in Budget Law allows for CLs for budgetary transfers to a State Agency Contingency Fund. • Chile – Methodology for estimating CLs in Government is published. The Government has established an Economic and Social Stabilization Fund and a Pension Reserve Fund (linked to copper prices and fiscal revenues/GDP ratio). Since it is difficult to measure contingent liabilities (CLs), a practical approach to monitoring them is to start with estimating the maximum liability (if this is an explicit contingent liability like a subnational loan that is guaranteed by the Central Government) and then decide how much of this fiscal risk to cover. (The examples of practices in Chile and Colombia are good in this regard.) In the case of explicit contingent 69 liabilities in Chile, such as student Loans, minimum guaranteed income in PPPs for Airports, roads and energy projects, bring contingent liabilities to “adjusted CLs” on the basis of Net present values (NPVs). When it comes to implicit contingent liabilities, the practice in some best practice countries has been to focus on doing fiscal sustainability scenario analyses, stress tests, and contingency planning for the potential fiscal costs of natural disasters and financial sector implicit liabilities only as a way to set up their framework for contingent liability management in the Government. They are not incorporated in the annual budgets of Government. Some countries do disclose in practice, but only aggregate information is published. For instance, Chile, Pakistan and the U.S.A. disclose the fiscal cost of past banking crises, and past costs of stabilizing fuel prices. However, even where govt. would not disclose implicit fiscal risks, it would be prudent to analyze exposure to implicit fiscal risks (as sometimes amounts can be very large).22 Conclusion Addressing a country’s high stock of “legacy” debt at the subnational level and making associated future subnational debt-service obligations more manageable in terms of prudent levels of cash flows is only one aspect of subnational debt restructuring. Underlying these efforts is the aim of achieving medium-term fiscal sustainability while, at the same time, better monitoring and minimizing fiscal risks to the government. Such fiscal surprises could emerge from the different financing vehicles that are often used by subnational governments (including municipalities) to finance their future investment and spending needs to achieve their equitable growth and development objectives. This paper draws from a rich history of cross-country experiences, and discusses how various countries emerged from their subnational indebtedness pressures. All these measures were undertaken as an integrated package that restructures the debt and implemented subnational fiscal adjustments over a medium-term period of implementation. These experiences from selected countries offer important lessons for China going forward. In the end, the design of subnational debt restructuring approaches, insolvency regimes, fiscal adjustment policies in subnational governments will have to be informed by a country’ s own history, circumstances, economic, legal and institutional structure and the external environment it faces when embarking on such a process of subnational restructuring. References Brixi, Hana Polackova and Mody, Ashoka (2002). “Dealing with Fiscal Risk”. In Government at Risk by Hana Polackova-Brixi and Allen Schick (eds.), World Bank, Washington D.C. 2002. Farvacque-Vitkovic, Catherine and Mihaly Kopanyi eds. (2014) “Municipal Finance: A Handbook for Local Governments” Washington D.C.: World Bank. Irwin, Timothy C. (2015). “Defining the Government’s Debt and Deficit”. Journal of Economic Surveys, Vol 29, No.4 pp. 711-732. Ministry of Finance, China (2015). Budget Report to the National People’s Congress, Beijing. 22 See R. Velloso, IMF, Fiscal Affairs Dept. “Good Practice in Fiscal Risks Disclosure: International Experience”. 70 OECD (2015), OECD. Fiscal Decentralization Database. 2015. Schaechter, Andrea, Tidiane Kinda, Nina Budina and Anke Weber (2012). Fiscal Rules in response to the Crisis – toward the “Next Generation”rules, A New Dataset. International Monetary Fund Staff Working Paper No. WP/12/187. July 2012. Vammalle, C and C. Hulbert (2013), “Subnational Finances and Fiscal Consolidation: Walking on Thin Ice”, OECD Regional Development Working Papers, 2013/02. OECD Publishing. 71 Lessons for Borrowing Policy and Subnational Risks Management Lessons from Country Experiences Sudarshan Gooptu Global Lead, Fiscal Policy, Macroeconomics & Fiscal Management (MFM) Global Practice, Equitable Growth, Finance and Institutions (EFI) Vice-Presidency Presentation at Ministry of Finance, P.R.China-World Bank Summit on Subnational Debt Management and Restructuring , Nanning, Guangxi Province, P.R. China. October 22, 2015. 72 Fiscal Adjustment and Subnational Borrowing Regulations Jorge Martinez-Vazquez and Violeta Vulovic23 I. Why are borrowing regulations of significance? The development literature widely acknowledges the importance of infrastructure for economic growth, quality of life and poverty reduction. 24 And with deep decentralization trends throughout all regions of the world, and with subnational governments on average being in charge of approximately two- thirds of total public infrastructure spending, there has been a natural increase in importance of subnational borrowing for financing this infrastructure.25 Even though there are some countries around the world that prohibit outright any borrowing by subnational governments,26 most other countries allow subnational borrowing because in their view the efficiency and equity benefits of borrowing outweigh the associated macroeconomic risks.27 But on the whole, factors such as the lack of institutional capacity and the history of sub-national government defaults in other decentralized systems give central governments substantial arguments to regulate sub- national government autonomy by introducing effective borrowing controls. The challenge is to simultaneously achieve the goals of providing borrowing autonomy and at the same time to preserve fiscal discipline by preventing the insolvency of sub-national governments and assuring national fiscal sustainability.28 One key aspect of the process is the presence of “moral hazard.” Subnational governments have less incentive than central governments to be concerned with the macroeconomic impact of their policies because they do not bear –or at least they perceive so-- the full cost of their actions. In short, subnational governments are not concerned—or as concerned-- with national fiscal sustainability as central governments are. 29 While well designed fiscal decentralization systems, especially on the side of 23 International Center for Public Policy, Georgia State University. October 2015. We are grateful to Yasin Civelek for excellent research assistance. The views expressed here are those of the authors alone. 24 See, for example, OECD (2006) and World Bank (1994). 25 See Martinez-Vazquez and Timofeev (2015). 26 That is the case in a good list of developing countries. Among developed countries, Denmark is among the few that have an outright prohibition. 27 The advocates of sub-national borrowing typically emphasize four potential benefits: (i) expansion of the sub- national fiscal space for the infrastructure financing; (ii) efficient and inter-generationally equitable outcomes from infrastructure financing through borrowing; (iii) increased fiscal transparency of the sub-national governments; and (iv) a deepening of national financial markets. Empirically, a positive effect of the availability of subnational borrowing on the provision of infrastructure service has been found (Freire and Petersen, 2004; Leigland, 1997; Peterson and Hammam, 1998). 28 Fiscal discipline requires imposing constraints on all three fiscal aggregates: total revenues, fiscal balance, and public debt (Fölscher, 2007). 29 Past macroeconomic crises involving public debt such as those in Russia, Argentina, Brazil, and East Asia, have brought up the importance of fiscal sustainability as an important component of macroeconomic stability. The more recent experience of peripheral European countries during the great recession that started in 2008 has made the link between fiscal sustainability and macroeconomic stability much more salient. 73 subnational revenue autonomy, can enhance or at least not harm fiscal sustainability (Fukasaku and De Mello, 1998), there is also consensus that decentralization can pose significant risks to fiscal sustainability and that a disciplined subnational borrowing process is much needed (Ter-Minassian, 1997b).30 Due to the potential long-term consequences of sub-national borrowing on fiscal sustainability and macroeconomic stability, most countries manage and supervise sub-national borrowing and debt by implementing ex-ante and/or ex-post borrowing regulations. The ex-ante regulations can consist of more or less direct control by the central government, of fiscal rules predetermined in the constitution or organic laws, or of a reliance on the financial markets and its mechanism to control borrowing. On the other hand, ex-post regulations consist of sanctions for non-compliance to the rules and for imprudent behavior. There is also consensus that both ex-ante and ex-post regulations should be used simultaneously, and should consider both the borrowers and the lenders (Webb 2004). One view in the literatures is that in regulating subnational borrowing it should be enough to rely on financial markets and the rules it imposes on debtors and creditors. Other legal rules are seen as unnecessary because market conditions already impose effective sanctions through higher interest rates and denying any lending. However, the history of sub-national borrowing in some decentralizing countries suggests that exclusive reliance on the financial markets in maintaining sub-national fiscal discipline may not be enough (Ter- Minassian and Craig, 1997). The necessary conditions of developed financial markets, availability of financial information, and no expectation of bailouts by the central government are generally not met, and defaults can have very long term negative consequences.31 A commonly accepted definition of fiscal sustainability states that the fiscal balance and the underlying trends are such that in a steady state the ratio of outstanding debt and debt servicing to GDP is not increasing over time (World Bank, 2010b). In a similar way, the IMF (2001) defines a set of fiscal policies as sustainable if a borrower is able to continue servicing its debt without an unrealistically large future correction to its income and expenditure. For the purpose of this paper we will define fiscal policy to be sustainable if the present value of future primary surpluses equals the current level of debt. More in particular, in this note we examine the international practices with subnational borrowing regulations and we advance the results form a more detailed statistical analysis examining whether any particular borrowing regulatory framework perform in a superior manner in maintaining fiscal sustainability, as well as what factors are important in choosing particular types of subnational borrowing regulations. Despite the importance of these issues, little systematic empirical work has been done so far on the effect of sub-national borrowing on fiscal sustainability. The existing literature does not offer a definitive answer on whether borrowing at the sub-national level should be allowed, and if so, how it should be regulated. The few cross-country empirical studies that evaluate these effects use either only some aggregate measure of borrowing autonomy that does not take into account different types of regulations, 30 The empirical literature on this issue is inconclusive (Martinez-Vazquez et al. 2015), but this is not surprising given that the outcomes are very much dependent on the decentralization system design and actual operation. 31 As an example, in the 1840s eight American states defaulted on their debts and these same states still continued paying a premium on their debt in the 1990s (English, 1996). 74 monitoring and enforcement, or focus only on the effect of the fiscal rules. Moreover, most of these past studies suffer from important econometric issues, including not addressing the potential reverse causality between fiscal sustainability and chosen types of borrowing regulations, not modeling a dynamic process in fiscal sustainability, or just focusing on the subnational rather than the general government fiscal performance. II. International experience with borrowing regulations As those who disagree with fiscal decentralization emphasize, giving more responsibilities to sub- national governments may endanger their fiscal sustainability and macroeconomic stability, suggesting that to maintain sustainability borrowing controls at the sub-national level are required. The literature on sub-national borrowing emphasizes higher government levels’ provision of an implicit guarantee to sub- national government debt as one of the main problems with borrowing at the sub-national level, causing a classical moral hazard situation. There are different ways in which a national government can contribute to prudent borrowing and these alternatives have been much debated. Figure 1. Broad types of ex-ante sub-national borrowing regulations (relative frequency in the sample*) 40% 35% 30% 25% 20% 15% 10% 5% 0% Prohibited Administrative Centrally-Imposed Self-Imposed Cooperative Market Discipline Rules Rules 1990 2000 2008 Source: Author's calculations; *Sample consist of 60 industrialized, developed and countries in transition. As Figure 1 presents, most of the countries that introduced borrowing at the sub-national level after 1990, preferred centrally-imposed rules or direct control by the central government as the dominant type of regulation. Furthermore, there has been a relative decrease in sole reliance on financial markets in regulating sub-national borrowing, which may be explained by experience gained from recent crises in which sub-national borrowing played major role. Moreover, in the last two decades, there has been an increased trend of imposing legal sanctions for non-compliance, mostly in case when sub-national 75 borrowing is dominantly regulated by centrally imposed rules (Figure 2). This trend of imposing legal sanctions for non-compliance is mostly due to those countries that have introduced borrowing at the sub- national level during this period, rather than to the changes in those that have already been present in the sub-national capital market (Figure 3). Most countries manage and supervise sub-national borrowing and debt by implementing ex-ante and/or ex-post borrowing regulations. Ex-ante regulations can consist of more or less direct control by the central government, of fiscal rules determined in the constitution or organic laws, or of reliance on the financial markets and their mechanisms. On the other hand, ex-post regulations consist of sanctions for non-compliance to the rules or for imprudent behavior. Webb (2004) contends that both ex-ante and ex-post regulations should be practiced simultaneously, and should consider both the borrowers and the lenders. Reliance on only ex-ante controls gives both the borrowers and the lenders incentive for irresponsible behavior since it bears no consequences. On the other hand, reliance on only ex-post regulations may give space to large sub-national governments to over-borrow and build up debts so large that the central government cannot enforce them to bear the consequences, given their importance in the national economy. Figure 2. Sanctions for non-compliance by type of ex-post sub-national borrowing regulations (relative frequency in the sub-sample*) 35% 30% 25% 20% 15% 10% 5% 0% 1990 2000 2008 Source: Author's calculations; *Sub-sample consists of 39 countries in 1990, 47 in 2000 and 51 in 2008 that allow sub-national borrowing. This chapter reviews the four main institutional settings that have been used to regulate the operations of sub-national credit markets. They represent the ex-ante regulations and sanctions for non- compliance, as an ex-post regulation of sub-national borrowing. The ex-ante regulations reviewed in this chapter include four broad types defined by Ter-Minassian and Craig (1997), namely, market discipline, fiscal rules, administrative and cooperative regulation. 76 THE EX-ANTE REGULATIONS Ex-ante regulations consist of ex-ante control and monitoring of sub-national borrowing and fiscal performance. These regulations specify the purpose, types, and procedures of sub-national borrowing. Liu and Waibel (2006) summarize the key elements of ex-ante regulations commonly used in practice. The first is allowing borrowing only for financing long-term capital investments, which is also known as the “golden rule”. The second type of the ex-ante regulations set limits on key fiscal variables, such as the primary and/or fiscal deficit, debt service ratio, etc. Finally, some frameworks include requirements for the sub-national governments to establish a medium-term fiscal framework and a transparent budgetary process. To improve fiscal transparency, more and more countries have introduced credit rating systems for sub-national governments, as an element of the regulatory framework for sub-national borrowing. Figure 3. Allowing borrowing at the sub-national level and imposing legal sanctions for son- compliance (relative frequency in the sample*) 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Sanctions Allowed borrowing Source: Author's calculations; *Sample consist of 60 industrialized, developed and countries in transition. MARKET DISCIPLINE In some countries, the government relies solely on the capital markets to regulate sub-national borrowing. Market discipline means that the financial markets are capable of sending appropriate signals to prevent a borrower from entering the “unsustainable area”, and borrowing is limited by lenders’ willingness to invest. Credit agencies, such as Standard and Poor’s, Moody’s, and Fitch, provide both the lenders and the borrowers in the market with information about the risk of default. There are, however, certain conditions that need to be satisfied for the private financial markets to be an effective control 77 instrument for sub-national borrowing. These include: (i) capital markets must be free and open; (ii) potential lenders must have available information about the borrower’s outstanding debt and repayment capacity; (iii) there should be no chance or possibility of a bailout of lenders by the central government; and (iv) borrowers must have the ability to respond with adequate policies to the signals sent by the market (Lane, 1993). In this sort of setting, sub-national governments generally have direct access to the financial markets to meet their borrowing requirements. Also, they independently decide how much and from whom to borrow, and on what to spend the borrowed money. For example, provinces in Canada may borrow for any purpose, whenever, wherever, and however they wish. There are neither internal nor external federal controls over provincial borrowing, and they do not even need to provide any information on their borrowing to the federal government (Bird and Tassonyi, 2001). Unlike provinces, municipalities face a very explicit hard budget constraint. Local borrowing requires prior provincial approval and is severely limited. Similarly to Canadian Provinces, Finish and Swedish municipalities do not need authorization from higher authorities to raise loans and can borrow from both domestic and foreign sources without any special conditions (Council of Europe, 1996b, 2009). Market discipline is only effective if the capital market is free and open. Restricted access to foreign capital markets limits the available options and creates a suboptimal financial sector portfolio (Giugale et al., 2000). There has been an increasing trend of allowing sub-national borrowing in foreign capital markets over the last two decades, but mostly only with an approval by the central government authority (Figure 4). Furthermore, availability of information and full transparency on outstanding debt and capacity to pay are essential to market discipline. However, obtaining reliable financial information, especially from the sub-national governments, often requires significant effort. Moreover, not all the sub-national governments follow a standardized accounting plan, hold uniform registers of their assets and liabilities, or publish information on debt and capacity to pay. In addition, hidden extra-budgetary funds weaken transparency. Additionally, moral hazard undermines the effectiveness of market discipline in checking sub-national governments’ excessive indebtedness. Bailouts encourage the expectation of future rescues and moral hazard type behavior of both the borrowers and the lenders. Finally, market signals, such as interest rates, can affect borrowers’ financial behavior in choosing more solvent fiscal policies. However, the borrowers must be sensitive to the market signals for market discipline to be effective, that is, the decisions about borrowing should change depending on the interest rate. However, in many parts of the world, capital markets at the local level are inadequately developed to be able to provide efficient discipline to sub-national governments. In such circumstances, credit rating agencies at the sub-national level are becoming increasingly important to evaluate the performance of intergovernmental systems. In this same context, some sub-national governments have adopted fiscal responsibility rules (that are self-imposed) trying to improve their credit ratings in the market. Examples of these trends are seen in Canada, Switzerland, and the United States. Some countries in Latin America, such as Argentina, Brazil, Colombia and Peru, recently have sought to follow this approach, at least partially, with the introduction of Fiscal Responsibility Laws (Webb, 2004). As mentioned above, the Canadian government relies solely on market discipline in controlling sub- national indebtedness. Credit rating companies evaluate sub-national creditworthiness. However, even 78 Canada’s fully-developed financial markets have not been fully able to control excessive indebtedness of the sub-national governments. In fact, in the mid-1990s, sub-national debt reached 23 percent of GDP (Bird and Tassonyi, 2001), (prompting) the provinces to adopt fiscal adjustments programs. Brazil and Argentina, without meeting all necessary market conditions, did in fact rely on some sort of market discipline approach in the 1980s, which had very unfortunate consequences. In Brazil, sub-national debt jumped from 1 percent of GDP in the early 1970s, to 20 percent in the mid-1990s, with five large federal bailout interventions (three for states and two for municipalities) (Bevilaqua, 2002). Market based sub-national borrowing regulations can take different forms. Dillinger (2003) compares the United States’ and the European model for market based mechanism and concludes that while the United States’ model relies primarily on municipal bonds, the European model relies dominantly on specialized banks to finance sub-national borrowing. However, municipal bonds are becoming more and more popular in Europe recently. The largest owners of municipal bonds in the United States are individual investors, mutual and money market funds, and the commercial banks. After being issued, municipal bonds can be sold in the secondary market, and are considered relatively safe from default, despite some opposite examples in the recent period. Some of the specialized banks in Europe are owned by the municipalities (e.g. Finland and Sweden), while others are founded by the national governments and have later been privatized (e.g. Dexia in France). RULE-BASED APPROACH Rule-based regulations consist of fiscal rules imposed by the central government and specified in the constitution or in the organic laws. Such rules introduce a constraint on fiscal choices by sub-national governments in order to guarantee that fiscal outcomes will remain predictable and robust regardless of the government in charge. Rules may take different forms: ceilings on debt or total borrowing, deficit targets, maximum expenditure rules, the “golden rule” (proceeds from borrowing must be spent exclusively on capital projects), or rules related to debt repayment capacity. Borrowing and debt ceilings represent the borrower’s upper legal limits of total indebtedness and are generally simple and easy to monitor. A deficit target has the advantage of simplicity and of being easily understood by the wider public, but it may be unsuccessful in preventing excessive debt accumulation because of the off-budget items. The most frequent deficit target rules are those targeting the overall budget deficit (e.g. Austria, Belgium, Spain, and most U.S. states) or the operating deficit (e.g., Norway). However, deficit target rules can also be met at higher levels of revenues and expenditures, which may have macroeconomic implications. Expenditure rules set the limits on the expenditure level, and are conceptually simple, easy to monitor, and can be most directly controlled. However, an expenditure limit can be more difficult to implement at the sub-national level than a deficit target and may not necessarily be able to prevent debt accumulation, since spending could be pushed below the line. Furthermore, the “golden rule”, limiting the sub-national governments’ borrowing to finance capital investment only, mostly satisfies the intergenerational equity justification for borrowing. However, borrowing for infrastructure does not guarantee by itself the macroeconomic and debt stability. Typically, infrastructure investments are required to provide “adequate” economic and social rates of return to be desirable or be approved. Many 79 countries currently implement some form of the “golden rule” (e.g. the United Kingdom, Germany, Spain, and most states in the U.S.). Finally, rules related to the capacity to repay debt attempt to stimulate the workings of the market discipline approach by relating the limits on indebtedness to expected debt service on the debt (e.g. Colombia and Hungary in the 1990s). These rules, however, might not be as effective in controlling debt accumulation if financial conditions are manipulated. Figure 4. Allowing sub-national borrowing in foreign capital markets (relative frequency in the sub-sample*) 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1990 2000 2008 Prohibited foreign Allowed foreign w/ approval Allowed foreign Source: Author's calculations; *Sub-sample consists of 39 countries in 1990, 47 in 2000 and 51 in 2008 that allow sub- national borrowing. Fiscal rules have the advantage of being generally transparent, more effective in addressing long-term sustainability and intergenerational equity, and relatively easy to monitor. They can, however, be counterproductive if poorly designed, or not adequately enforced. Most countries using the rule-based approach use a variety of rules, some of which are redundant. Main disadvantage of the rule-based approach is the trade-off between ensuring compliance and preserving flexibility. Strict fiscal rules leave little room for adjustments in case of unexpected economic downturns, while more flexible fiscal rules lack credibility and may fail to impose sufficient discipline. In practice, the efficacy of fiscal rules for sub- national governments primarily depends on the ability to monitor the debt. There has been particularly increased trend to impose limits on sub-national debt and borrowing during the last two decades (Figure 5). The use of the “golden rule” has also increased, but not by as much. All but one state in the United States (Vermont) has a balanced budget requirement. Budget rules vary significantly across the U.S. states, mostly applying only to the operating budget (general fund). In addition, as of 2008, 30 states also operate under tax or expenditure limitations (Waisanen, 2008). Several studies investigate the effectiveness of sub-national government rules in the context of the U.S. states. 80 Most authors conclude that rules do enforce some budget discipline on the U.S. states, in terms of lower deficits and quicker reaction to negative fiscal shocks (Poterba, 1994; Alesina and Bayoumi, 1996; Poterba and Von Hagen, 1999; Poterba and Rueben, 1999). In the European Union, within the Stabilization and Growth Pact, limiting the overall level of public debt as well as annual total budget deficits, raises the question about whether the debt limit should be shared among the levels of government. In most countries it is assumed that the central government should be responsible for the overall limit of public debt. Public debt is much lower at the sub-national compared to the central government level, being just above 8 percent of total debt in Germany to around 19 percent in Switzerland (Swianiewicz, 2004). In most European Union countries the ratio of the sub- national debt to GDP is pretty low, on average around 5 percent. The only “outliers” are the Netherlands and Spain with over 8 percent of the sub-national debt to GDP. In Belgium, only the central government is responsible for complying with the European Union fiscal rules, but with the agreements set between the central and the sub-central levels of government, the commitments for complying with these constraints is shared among all levels of government. Figure 5. Imposing limits on borrowing and debt and the “Golden Rule” (relative frequency in the sub-sample*) 80% 70% 60% 50% 40% 30% 20% 10% 0% 1990 2000 2008 Limits on Debt and Borrowing The "Golden Rule" Source: Author's calculations; *Sub-sample consists of 39 countries in 1990, 47 in 2000 and 51 in 2008 that allow sub- national borrowing. Switzerland’s approach to the sub-national borrowing regulation is an example of self-imposed fiscal rules. 26 Swiss cantons apply different regulations which are set in each Canton’s law. In many cantons, borrowing is allowed only for financing capital expenditures, and if the local and/or cantonal government has the financial capacity to pay the interest on debt as well as the amortization out of the current budget. Dafflon (2002a) discusses the sub-national borrowing regulation practice in the Fribourg canton where for each project that cannot be financed from the current revenues, then the borrowing for its financing requires the cantonal approval. 81 ADMINISTRATIVE APPROACH The administrative approach is completely the opposite from the market discipline approach, giving the central government direct control over sub-national borrowing. It may take different forms, such as setting an annual or even more frequent limits on the overall sub-national government debt; prohibiting external (foreign) borrowing; reviewing and approving individual borrowing operations (including approval of the terms and conditions); or centralizing all government borrowing with on-lending to the sub-national governments. The approval of each borrowing issuance requires an evaluation of the financial terms and conditions under which each operation is contracted. The administrative approach is more frequently used by unitary countries and less by (federal type countries.) Direct involvement of the federal government in micromanaging each credit operation at the sub- national government level represents one of the disadvantages of this approach, since it is the opposite of the fiscal decentralization idea. Moreover, this approach may unnecessarily increase federal bureaucracy and cause undesirable inefficiencies in the financial system, and may even be incompatible with a country’s Constitution if it allows the sub-national government free access to the capital market. However, a major disadvantage of this approach is the moral hazard resulting from the fact that the central government may find it difficult to refuse to financially support the lower levels of the government in the case of impending defaults. On the other hand, the administrative approach has several advantages. First, the central government can control both the macroeconomic and the external debt policy. Second, the central government’s control may increase the sub-national borrower’s credibility, given that the foreign lenders often require a central government guarantee, and it may result in better terms and conditions received in the foreign financial markets. Countries like Denmark, Greece, Ireland, Mexico, and the United Kingdom practice the administrative control approach in regulating sub-national borrowing. In Mexico, the states and municipalities, including their decentralized agencies and public enterprises, can only borrow domestically to finance investment outlays up to the ceilings set by their respective legislatures. Unlike several other countries in Latin America, Mexico does not have a Fiscal Responsibility Law even under consideration. It uses financial sector regulations instead to motivate state-level prudence. In the United Kingdom, a local authority may not, without the consent of the Treasury, borrow from a lender from abroad or in a currency other than sterling. In Spain, for example, foreign debt and bond issuances by the sub-national governments are subject to the approval of the Ministry of Finance. During the 1980s, Australia centralized regulation of sub-national borrowing through the Loan Council, but this direct control system did not turn out to be effective, and now the sub-national governments are free to access the capital markets directly. The functions of the Loan Council were restructured in the mid 1990s, and excessive indebtedness is now cooperatively controlled (Craig, 1997; Dillinger, 2003; Koutsogeorgopoulou, 2007). Denmark provides an interesting example of the administrative approach to sub-national borrowing regulation. In general, sub-national borrowing in Denmark is prohibited, but in some cases this rule is waived. Permission for borrowing issuance, for which the municipalities apply individually, is granted if the overall borrowing ceiling has not been exceeded and if the municipality’s debt does not exceed 30% of total municipality’s expenditures. The borrowing and debt ceilings are negotiated annually with local 82 government associations. Furthermore, the general rule is that, if borrowing is permitted, both current and capital budgets need to be balanced. Nevertheless, during the 1990s between 40% and 80% of Danish municipalities’ deficits) were financed through borrowing, resulting in local debt of 4.5% of GDP in 1998 (Jorgen and Pedersen, 2002). Similarly, the United Kingdom applies an administrative approach to sub- national borrowing regulations, but in the British case the borrowing limits differ among sub-national governments (Watts, 2002). Limits are allocated to the local governments depending on their specific needs for housing, education, etc. Allocations are increased or decreased based on the efficiency and effectiveness of the local governments and can be adjusted for special needs (Dafflon, 2002b). COOPERATIVE APPROACH Under this approach, the sub-national borrowing controls are designed through a negotiation process between the federal/central and the lower levels of government. The sub-national governments are actively involved in reaching an agreement on overall general government deficit targets, on the main revenue and expenditure items, as well as on the limits on financing of the individual sub-national jurisdictions. This approach is in practice in some European countries and in Australia. In Austria, for example, the “Consultation mechanism” between different levels of government and the Stability and Growth Pact were implemented in 1999 (Thöni, Garbislander, and Haas, 2002) to ensure lowering and maintaining the overall deficit below 3 percent of GDP. Similar arrangements exist in Spain (Laborda et al., 2006). In Belgium, sub-national borrowing is supervised by a High Finance Council (HFC), which is comprised of members nominated by the federal, regional, and community levels, and the Belgian National Bank. In Australia, a fiscal institution called the Loan Council coordinates the fiscal policies and borrowing decisions of the Australian states. The cooperative approach combines many individual advantages of the other three approaches, which is both its main strength and its main weakness. A clear advantage lies in promoting dialogue and the exchange of information across various government levels, as well as in raising awareness of the macroeconomic implications of their budgetary choices. However, in order to be effective, this approach requires the central government to be strong and able to effectively guide the intergovernmental negotiations, which in many emerging markets may not be the case (Joumard and Kongsrud, 2003). The main weakness of this approach is that, because it combines components of other three approaches, when it is poorly implemented it reproduces the flaws of other approaches, instead of their advantages (Ahmad, et al., 2005). As already mentioned above, since the Loan Council’s functions were restructured in the mid-1990s, sub-national borrowing in Australia has been cooperatively controlled. Jurisdictions are required to submit their total financial requirements for the upcoming year to the Loan Council with no requirement for submitting specific project details. Then the Loan Council evaluates these nominations with regard to the jurisdictions’ fiscal position, the infrastructure needs and the macroeconomic implications of borrowing. In the event when the Loan Council has concerns about certain nominations, it has the right to request the jurisdiction to justify the nomination, and if needed, it can amend its fiscal strategy. So far, the restructured Loan Council, complemented by the financial markets and rating agencies, has been 83 successful in controlling sub-national fiscal behavior (Craig, 1997; Koutsogeorgopoulou, 2007; Webb, 2002). A key role in managing sub-national borrowing in Belgium is played by the “Public Sector Borrowing requirements” in the High Finance Council (HFC). The HFC is composed of academics, members of the National Bank of Belgium and the representatives of all levels of governments. The committee monitors and analyzes the borrowing requirements of all levels of government at regular intervals and, based on a concept of sustainability, formulates recommendations about the medium and long-term budgetary targets for the different government levels. Based on the HFC’s recommendations, the agreements between the central government and the regions and the communities are formulated, covering a period of five to six years and committing the sub-national governments to meeting specific annual budgetary targets in terms of their borrowing requirements. In order to ensure that public finances are consistent with the budgetary targets, municipalities are subject to the “golden rule” under which deficits are only allowed for investment. On the recommendation of the HFC, the central government can limit the borrowing capacity of a non-compliant region or community to prevent endangering economic stability or the external balance. So far, however, the HFC has not considered it necessary to use this sanction on any of the regions or communities (OECD, 2007). According to Liebig, et al. (2008), the sub-national borrowing regulation in South Africa is a combination of the cooperative and the marked based approach. The cooperative component originates in the South African Constitution where Article 3 requires a “co-operative government”. Furthermore, different spheres of the government control each other in terms of who borrows how much. On the other hand, the South African legal setting for sub-national borrowing is also partly market-based, since the sub- national entities can generally borrow as much as they want. The municipal councils authorize borrowing issuances and there are no country-wide debt limits. EX-POST REGULATION As already pointed out, the effectiveness of ex-ante regulations is limited without an ex-post mechanism for dealing with sub-national insolvency. Even though ex-ante regulations are very important for minimizing the risk of defaults, they cannot prevent them in all cases. Sub-national insolvency may occur because of sub-national fiscal and debt mismanagement but also because of external shocks. Ex-post control mechanisms consist of a set of predetermined rules for allocating the default risk. They provide a basis for both borrowers’ and lenders’ expectation that in case of insolvency, they both would share the burden. Properly designed ex-post regulations enforce the hard budget constraint on sub-national governments. Countries generally apply two main approaches in ex-post regulation of the sub-national borrowing, namely the judicial and the administrative approach. The judicial approach involves the courts which make key decisions and give guidance on the restructuring process. The advantage of the judicial approach is that it neutralizes political pressure. However, the ability of courts to impose fiscal adjustments on sub- national governments is very limited. The administrative approach, however, often allows political intervention of the higher levels of government in resolving the sub-national insolvency. 84 Depending on the factors, such as history, political and economic structure, etc. countries apply various approaches for ex-post regulation of the sub-national borrowing. For example, Hungary and Brazil apply the administrative approach, while South Africa and the United States prefer a combination of the judicial and the administrative approaches. Moreover, there is a uniform approach across states in the United States for dealing with municipal distress. Any ex-post control mechanism consists of three central elements. The first is the definition of insolvency that acts as a procedural trigger. Different countries define insolvency. While Hungary and the United States define insolvency as inability to pay debt, South Africa uses one definition for serious financial problems and another for persistent violation of financial commitments. The second element is the debtor’s fiscal adjustment to bring in line spending with revenues as well as borrowing with capacity to service debt. Even when the sub-national governments have significant autonomy in controlling expenditures and raising revenues, fiscal adjustment often requires difficult political choices of reducing spending and raising revenues. Finally, the third one includes negotiations between the debtor and creditor to restructure debt obligations. In case of the administrative approach, the higher government level tends to restructure sub-national debt into longer-term debt instruments, which was the case in Brazil in 1997. However, the debt discharge is typically limited to the judicial approach (Liu, 2008). In summary, there is a wide variety of both ex-ante and ex-post sub-national borrowing regulations that countries implement. The regulations reflect the level of development of the financial markets, the political power of different levels of government, and macroeconomic and fiscal conditions. Each type of sub-national borrowing regulation has both advantages and disadvantages, which determine how suitable each is for a particular country’s circumstances. Reliance on only ex-ante controls gives both borrowers and lenders the incentive for irresponsible behavior since they bear none of the consequences. On the other hand, reliance on only ex-post regulation may give space to large sub-national governments to over- borrow and build up such large debts that the central government cannot enforce them to bear the consequences given their importance in the national economy. Finally, as a county’s circumstances change over time, the country may change its preferred mechanism to control sub-national behavior in financial markets. III. Advancing the main results on the selection of borrowing rules and on the impact of rules on fiscal sustainability For the empirical analysis we use unbalanced panel data for 57 industrialized, developing, and transition countries between 1990 and 2008. Two alternative dependent variables are used; namely, the primary balance32 at the general government33 level and at the sub-national level.34 The main variables of interest are four broad types of sub-national borrowing regulations, first categorized by Ter-Minassian and Craig (1997); namely, market-based, rule-based35, cooperative, and administrative regulation. The 32 That is, Revenues minus (Expenditures net of Interest Payments). 33 The general government sector consists of entities that fulfill the functions of government as their primary activity and can be divided into central, state, and local government subsectors, depending on a country (IMF, 2001: p.33). 34 The sub-national government represents all levels of government below the central government level. 35 With a distinction made between centrally-imposed and self-imposed rules. 85 results obtained from using these types of sub-national borrowing regulations are compared with those obtained from prohibiting borrowing at the sub-national level altogether. To evaluate the relationship between sub-national borrowing regulations and fiscal performance at the general and the sub-national government levels, two methodologies are applied. First, we use the “system” GMM estimator to evaluate how the primary balance changes as a result of changes in the level of sub-national outstanding debt, sub-national borrowing regulations, revenue and expenditure autonomy, and in control variables. Second, we apply duration analysis to investigate the effect of the main and control variables on fiscal sustainability, when alternative levels of the primary balance are defined as fiscally sustainable. Finally, we address the potential reverse causality problem between the primary balance and the types of sub-national borrowing regulations by applying the multinomial logit approach in the first state to estimate the predicted probabilities of choosing each type of regulations. This methodology allows the investigation of potential determinants of choosing each of the sub-national regulation types at the same time. Advancing our main results, first, concerning the selection of rules, we find that the depth of the financial market is particularly important when choosing cooperative regulations and regulations based on centrally and self-imposed rules. Also, countries with higher primary balances (both at the general and subnational levels of government) are more likely to choose self-imposed rules and market-based regulations over the other types. On the other hand, countries with higher sub-national outstanding debt seem to be more likely to regulate the sub-national borrowing. The institutional design and history of the fiscal decentralization system has some effects on fiscal sustainability. The presence of subnational tax autonomy contributes to an increase in the general government primary balance but at the subnational level tax autonomy is on the margin not significantly high. In those countries with a history of subnational government bailouts primary balances on average are lower at both the sub-national and the general government levels than in other countries. On the effectiveness of borrowing regulations, we find that the “golden rule” (borrowing is only for capital investment purposes) and limits on debt and borrowing positively affect the primary balance at all levels of government. However, on the question of which regulations are most effective, we find that none of the broad types of sub-national borrowing regulations seem to have a significant direct effect on the narrow definition of fiscal sustainability at the subnational level. This is somewhat of a surprising result, given the amount of discussion and effort that has gone on in shaping up the different regulations. This negative result shifts the focus to what the impact may be of the different regulations on the overall fiscal balance of the country through the impact of the different fiscal behavior of subnational governments. And here there are some salient results. First, we find that cooperative type of sub-national borrowing regulations seems to have a positive effect on improving general government fiscal performance even in the case of high levels of sub-national debt and high dependence on subnational governments on intergovernmental transfers. In addition, the threshold level for fiscally sustainable primary balance appears to be quite determinant on how effective the different rules are: (i) When a particular level of the primary balance is predetermined as sustainable, 86 then self-imposed rules seem to be the only effective method in maintaining the primary balance above that threshold, for all government levels; (ii) more centralized types of regulations (administrative and cooperative) seem to be more effective with a higher threshold for fiscally sustainable primary balance, while more decentralized regulations (self-imposed rules) seem to show better performance with lower thresholds. References Ahmad, E., Albino-War, M., and Singh, R. (2005). Sub-national Public Financial Management: Institutions and Macroeconomic Considerations. IMF Working Paper Series 05/108. Alesina, A. F., and Bayoumi, T. (1996). The costs and benefits of fiscal rules: evidence from U.S. states. NBER Working Paper No. 5614. Bevilaqua, A. S. (2002). State government bailouts in Brazil. Inter-American Development Bank, Research Network Working Paper #R-441. Bird, R. M., and Tassonyi, A. T. (2001). Constraints on provincial and municipal borrowing in Canada: markets, rules, and norms. Canadian Public Administration, 44(1), 84-109. Council of Europe. (1996b). Structure and Operation of Local and Regional Democracy: Sweden. Council of Europe. (2009). Structure and Operation of Local and Regional Democracy: Finland. Craig, J. (1997). Australia. In T. Ter-Minassian (Ed.), Fiscal Federalism in Theory and Practice. Washington, D.C.: International Monetary Fund. Dafflon, B. (2002a). Capital Expenditure and Financing in the Communes in Switzerland. In B. Dafflon (Ed.), Local public finance in Europe: Balancing the budget and controlling debt. Cheltenham-Northampton: Edward Elgar Publishing. Dafflon, B. (2002b). The Theory of Sub-national Balanced Budget and Debt Control. In B. Dafflon (Ed.), Local public finance in Europe: Balancing the budget and controlling debt. Cheltenham-Northampton: Edward Elgar Publishing. Dillinger, W. (2003). Regulations and Markets: Brazil's Efforts to Control Subnational Borrowing. USAID. English, W. B. (1996). Understanding the costs of sovereign default: American state debts in the 1840's. American Economic Review, 86(1), 259-275. Fölscher, A. (2007). Local Fiscal Discipline: Fiscal Prudence, Transparency, and Accountability. In A. Shah (Ed.), Local Budgeting. Washington D.C.: The World Bank. Freire, M., and Petersen, J. E. (2004). Subnational capital markets in developing countries: from theory to practice. Washington D.C. / New York: The World Bank and Oxford University Press. Fukasaku, K., and De Mello, L. R. (1998). Fiscal Decentralization and Macroeconomic Stability: The Experience of Large Developing and Transition Economies. In K. Fukasaku and R. Hausmann (Eds.), Democracy, Decentralization and Deficits in Latin America. Paris: OECD. Giugale, M., Trillo, F. H., and Oliveira, J. C. (2000). Sub-national borrowing and debt management. In M. Giugale and S. B. Webb (Eds.), Achievements and Challenges of Fiscal Decentralization: Lessons from Mexico. Washington, D.C.: The World Bank. 87 IMF. (2001). Manual on Fiscal Transparency. Washington D.C.: International Monetary Fund. Jorgen, N., and Pedersen, M. (2002). Local Government and Debt Financing in Denmark. In B. Dafflon (Ed.), Local public finance in Europe: Balancing the budget and controlling debt. Cheltenham-Northampton: Edward Elgar Publishing. Joumard, I., and Kongsrud, P. M. (2003). Fiscal relations across government levels. OECD Economic Studies, 1(36). Koutsogeorgopoulou, V. (2007). Fiscal relations across levels of government in Australia. OECD Economics Department Working Papers No. 541. Laborda, J. L., Martinez-Vazquez, J., and Escudero, C. M. (2006). The Practice of Fiscal Federalism in Spain. International Studies Program Working Paper 06-23. Leigland, J. (1997). Accelerating municipal bond market development in emerging economies: an assessment of strategies and progress. Public Budgeting and Finance, 17(2), 57-79. Liebig, K., Bahrinipour, M., Fuesers, L., Knödler, B., Schönhofen, C., and Stein, M. (2008). Municipal Borrowing for Infrastructure Service Delivery in South Africa: A Critical Review: Deutsches Institut für Entwicklungspolitik. Liu, L. (2008). Creating a Regulatory Framework for Managing Subnational Borrowing. In J. Lou and S. Wang (Eds.), Public finance in China: reform and growth for a harmonious society (pp. 171–190). Washinton D.C.: The World Bank. Liu, L., and Waibel, M. (2006). Subnational Borrowing Notes for Middle-Income Countries. The World Bank. Martinez-Vazquez, J and A. Timofeev (2015) “Capital infrastructures and equity objectives in decentralized systems” in J. Frank and J Martinez-Vazquez (eds.) Decentralization and Infrastructure in the Global Economy: From Gaps to Solutions, Routledge, London and New York. OECD. (2006). Promoting pro-poor growth: infrastructure. DAC Guidelines and Reference Series. OECD. (2007). OECD Economic Surveys: Belgium. Peterson, G., and Hammam, S. (1998). Building local credit systems. Washington D.C.: The World Bank. Poterba, J. M. (1994). State responses to fiscal crises: The effects of budgetary institutions and politics. Journal of Political Economy, 102(4), 799-821. Poterba, J. M. (1995). Balanced budget rules and fiscal policy: Evidence from the states. National Tax Journal, 48, 329-329. Poterba, J. M., and Rueben, K. (1999). State Fiscal Institutions and the US Municipal Bond Market. In J. M. Poterba and J. v. Hagen (Eds.), Fiscal institutions and fiscal performance. Chicago: The University of Chicago Press. Poterba, J. M., and Von Hagen, J. (Eds.). (1999). Fiscal institutions and fiscal performance: University of Chicago Press and National Bureau of Economic Research. 88 Swianiewicz, P. (2004). The Theory of Local Borrowing and the West-European Experience. In P. Swianiewicz (Ed.), Local government borrowing: risks and rewards: a report on Central and Eastern Europe. Budapest, Hungary: DFID and Open Society Institute. Ter-Minassian, T. (1997a). Decentralization and Macroeconomic Management. IMF Working Paper WP/97/155. Ter-Minassian, T., and Craig, J. (1997). Control of subnational government borrowing. In T. Ter-Minassian (Ed.), Fiscal Federalism in Theory and Practice (pp. 156-172). Washington D.C.: International Monetary Fund. Thöni, E., Garbislander, S., and Haas, D.-J. (2002). Local Budgeting and Local Borrowing in Austria. In B. Dafflon (Ed.), Local public finance in Europe: Balancing the budget and controlling debt. Cheltenham- Northampton: Edward Elgar Publishing. Waisanen, B. (2008). State Tax and Expenditure Limits. The National Conference of State Legislatures. Watts, P. A. (2002). Local Government Capital Expenditure in England. In B. Dafflon (Ed.), Local public finance in Europe: Balancing the budget and controlling debt. Cheltenham-Northampton: Edward Elgar Publishing. Webb, S. B. (2004). Fiscal responsibility laws for subnational discipline: the Latin American experience. The World Bank Policy Research Working Paper No. 3309. World Bank. (1994). World Development Report 1994, Infrastructure for Development. Washington D.C.: World Bank. World Bank. (2010b). Public Expenditure Trends and Fiscal Sustainability. Afghanistan Public Expenditure Review 2010. 89 P.R. China Ministry of Finance-World Bank High-level Summit on SUBNATIONAL DEBT MANAGEMENT and RESTRUCTURING. October 22, 2015. Nanning, Guangxi Province, P.R. CHINA Fiscal Adjustment and Subnational Borrowing Regulations JORGE MARTINEZ-VAZQUEZ AND VIOLETA VULOVIC 90 SECTION IV - Roundtable Discussion on Options and Subnational Reform Priorities for China - Learning from Cross Country Lessons The participants in the roundtable discussion were Dr. Satu Kahkonen (Senior Director, World Bank), Professor Jorge Martinez-Vazquez (Professor, Georgia State University), Professor Roy Bahl (Professor, Georgia State University) and Professor Qiao Baoyun (Dean of China Academy of Public Finance and Public Policy, Central University of Finance and Economics). The roundtable was moderated by Dr. Bert Hofman (Country Director for China-Korea-Mongolia, World Bank). 91 Round-table Discussion: “Options and Subnational Reform Priorities for China - Learning from Cross Country Lessons” This section summarizes the discussion and key messages that emerged among panelists on their views about the main priorities for subnational debt-related reforms in China going forward. There was broad consensus among the panelists that:  The recent reform of the Chinese budget law in late-2014 was an important step towards putting in place a modern subnational debt management framework in China. For years, without formal on-budget debt-raising authority, Chinese local authorities have resorted to financing local public investment by borrowing through off-budget vehicles, such as Urban Development Investment Corporations (UDICs). The new budget reform measures require local governments to bring these off-budget liabilities explicitly on to the budget when making any fiscal policy decisions, including any maturing local government debt that is swapped with Central Government bonds. Provinces will now have an explicit quota for new on-budget borrowing. Since interest rates for government bonds are generally lower, this reform will reduce the debt servicing costs of local governments. Subnational government budgets will become more transparent.  During a transition phase, quotas on new provincial borrowing and close supervision of local government investment plans by the central government are probably appropriate. Chinese financial markets, and especially local banks which are often connected to local governments, are not yet sophisticated enough to effectively constrain local government borrowing and impose market discipline. Further devolution of borrowing and investment decisions, subject to fiscal rules, requires a clearer separation of responsibilities and budgets at the various levels of government. It will also require to build further on recent improvements in the fiscal planning capacity of subnational government entities. In the short-run, the central government will therefore need to retain substantial administrative control over local government budgets, debt management and investment plans.  Ensuring sound underlying local public finances is a prerequisite for further institutional reform. In the short run, the priority must be to ensure the sustainability of local government budgets. Sharp growth in local government liabilities following the fiscal stimulus program enacted by the Chinese government during the Global Financial Crisis (2007-2009), coupled with a recent slowdown of the Chinese economy, have raised doubts about the health of local government finances. A comprehensive debt sustainability analysis is therefore needed at all subnational entities. This analysis should take into account some of the key long-term challenges facing Chinese budgets, which are too often ignored, including unfunded pension obligations and the rapidly increasing costs of maintaining public infrastructure. Where required, fiscal consolidation and debt restructuring actions should be taken as soon as possible to bring local government budgets back onto a sustainable path. Given potentially limited scope for recurrent expenditure consolidation, this may require exploring further sources of local revenues, a cost-benefits analysis of public investment projects and assistance from higher levels of government. 92  In the medium-run, further evolution towards a rules-based fiscal framework could be considered. As local government budget planning becomes more sophisticated and local financial markets gain in depth, further decentralization of control over local budgets may become efficient. Local governments may be best placed to determine local expenditure needs and appropriate local financing means. Subnational ownership, accountability and funding may increase local pressure for efficient public investment projects. Relaxing central control can also reduce local incentives to circumvent formal regulation through opaque “back door” transactions and thereby increase budget transparency. Overall, future reform steps may consider further increasing the autonomy of subnational entities over their budget and debt management decisions, subject to broad fiscal rules to ensure sustainability of local borrowing.  A clear delineation of budget responsibilities across layers of government will be required. Expenditure responsibilities across different levels of government are still not clearly defined. Tax revenues are shared between central and subnational government, but local authorities have limited scope for setting their own tax rates and tax revenue sources. In practice, subnational entities conduct around 2/3 of expenditure, but receive about 1/2 of revenues. The gap is closed by transfers from the central government and subnational borrowing. For subnational entities to gain greater responsibility over their fiscal decisions, their budgets need to first be more clearly delineated. This implies that some expenditure responsibilities need to be reassigned. One panelist recommended, for example, a reassignment of expenditure responsibilities for most of social security and income maintenance programs away from the county governments and moved up to the Central and Provincial governments.  Need to introduce more sustainable sources of subnational revenues. A clear delineation of budgets will need to match expenditure responsibilities with revenue raising authorities at the local level. This will also improve the ability of subnational entities to manage their debt. In any such rebalancing effort, land-lease revenues will likely continue to be an important source of revenues for local budgets. In addition, several other taxes are currently under discussion as candidate sources of local revenue. A recurrent tax on the value of property is already being piloted in some regions of China. However, the implementation of such a tax is technically demanding, including the valuation of properties. Panelists noted that other taxes could include a vehicles tax, a vehicles fuel tax or a local piggy-back tax on an existing income tax base.36  The budget planning capabilities of subnational entities need to be enhanced. Multi-year budget planning will need to be fully implemented at the subnational level. Budget planning should also become more comprehensive, with full coordination of the recurrent and investment budgets. Economic growth targets and fiscal targets need to be set jointly and need to be mutually consistent, requiring coordination among government departments. The data quality, 36 A potential cost of moving tax-raising authority to the local level may be that the central government will find it harder to quickly mobilize funds, such as for support after regional natural disasters or for fiscal stimulus to stabilize the economy. 93 transparency and comprehensiveness of subnational budget data should be enhanced further, including the full recognition of off-budget state-owned enterprises and contingent liabilities. Overall, it was noted that these medium-term reforms would need to be introduced gradually over a period of at least five years. International experiences presented at this workshop suggest that reforms of the formal fiscal framework can only be successful if there is accompanying “cultural change” among all stakeholders towards more prudent public finance, debt and fiscal risks management. In addition, implementation capacity and human capital constraints need to be quickly identified and efforts to improve them in local budget authorities should be supported in earnest. To this end, the Central Government can assist subnational authorities, including by incentivizing subnational reforms through conditional financial assistance or conditional further debt swaps. Panelists highlighted that the World Bank Group—through the materials presented here, its convening power to bring the state-of the art thinkers and experienced practitioners from around the world, and its in-house expertise—has a wealth of international experience in the design and implementation support of appropriate reforms of subnational debt management frameworks in its client countries. The ongoing cooperation between the Chinese Government and the World Bank in two pilot subnational debt-related reform projects in Chongqing and Hunan are significant partnerships in this direction. 94 SECTION V - Concluding Remarks Mr. Chen Xinhua, Deputy Director-General, Budget Department, Ministry of Finance, China Dr. Satu Kahkonen, Director for Asia and Europe, Macroeconomics and Fiscal Management Global Practice, EFI Vice-Presidency, World Bank Mr. Xu Xiang, Deputy Director-General of Human Resources Development Center, Ministry of Finance, China 95 Chen Xinhua Deputy Director-General, Budget Department, Ministry of Finance, Government of China Respected Ms. Satu Kahkonen, honored specialists, ladies and gentlemen, time always passes so quickly when we enjoy sharing knowledge. After a full day of lively discussion, the international workshop on local government debt management is about to end, I, hereby, on behalf of the Budget Department of Ministry of Finance express the appreciation and thanks to the excellent speeches of specialists from home and abroad, the active participation of local government debt managers, and the warm hospitality of Department of Finance of Guangxi Zhuang Autonomous Region. Raising loans are a necessity for the local governments to carry out the process of industrialization and urbanization. Rational and moderate borrowing is helpful in improving the economic development. However, raising excessive loans and monitoring gaps could lead to debt crisis, which will affect the sustainable achievement of our economic development. This goal has been an important subject for many countries to study how to use the double-edged sword to effectively make full use of the borrowing through standardizing the activities of local governments, and successfully managing their debt size and risks. Specialists here have conducted full discussions based on their own research and practices, and the excellent speeches of many honored guests have left a deep impression on us. For example, Director Mr. Liu analyzed the paths and future of improving debt management in China from the perspectives of connection between debt management and budget management, as well as the national and local governance. Professor Qiao gave an in-depth analysis with focus on the taxation responsibilities, assignment of expenditure responsibilities, and development of the bond market in China, which summarizes the current characteristics of local debt system and reform logic in China. Ms. Lili Liu introduced the classic cases in New York and Detroit combined with the U.S. Chapter 9 Bankruptcy Code, pointing out that the size of the debt is not the only indicator of risk evaluation and it is necessary to also establish the connection between the ex-post and ex-ante management. Mr. Fernando Blanco compared the characteristics between Brazil and Mexico. Professor Jorge Martinez-Vazquez introduced the development of the local finance and debt in Russia, pointing out that it is necessary to maintain the balance between income and expenditure to prevent the increase of local debts. Ms. Abha Prasad pointed out that different strategies are adopted for the governance system with diverse political effects in different development stages, which makes the comparison more informative. Mr. Sudarshan Gooptu pointed out that the sustainability of fiscal rules played an important role in avoiding fiscal issues. Professor Jorge Martinez-Vazquez carried out an experimental analysis on the monitoring of subnational market rules, cooperation and administration. An open discussion was conducted by specialists from home and abroad and interaction is done with the audience, so as to share the latest theoretical results, provide valuable management experience, and put forward rewarding suggestions on the policies of subnational debt management in China. In light of this information, I think that there are common principles and universal practices for strengthening subnational debt management. The intervention and management of the central government, open and transparent borrowing mechanism, effective control of the size of debts, 96 standardized legal restraints and budget management were seen to be indispensable for U.S.A., European Union countries and developing countries such as India and Brazil. Furthermore, the stages of development of the management systems differ in various countries and regions. The objectives for the management of the creditor rights set by local governments are different as well. The evaluation of the outcomes and results of their responsibilities is different too. For example, the concept of emergency aid is different in Mexico. Ireland and Greece have different capacities in executing financial supervision. Obviously, the debt management of a country shall be based on its actual conditions. It should draw on the world-best practice management experience, but country-specific aspects should be taken into account. According to the Chinese practices, the debts of local governments accumulated over the years played a positive role in fighting international financial crisis and natural disasters, improving the people’s livelihood, protecting ecological environment and promoting economic and social development. Of course, there exist some problems such as nonstandard management, high financing cost and unreasonable structure. In recently years, the Chinese government has taken effective measures to strengthen subnational debt management with a strong sense of responsibility. In August 2014, the budget law was adopted by the National People's Congress Standing Committee. In September 2014, the State Council printed and distributed the opinions on strengthening subnational debt management which specified relevant requirements such as local governments’ debt size control body and method. Relevant departments of the Ministry of Finance have done a lot of work in these areas of budget management, bond issuing and risk control. We are soberly aware that improving subnational debt management is a long-term task. New circumstances, new problems and new challenges will be encountered. We will actively absorb the workshop results and constantly improve the subnational debt management. We also hope to deepen the cooperation with the World Bank in many aspects, so as to jointly advance subnational debt management theory and practice of our local governments. Finally, wish you all a happy journey home. Thanks! 97 Satu Kahkonen Director for Asia and Europe, Macroeconomics and Fiscal Management Global Practice Equitable Growth, Finance and Institutions (EFI) Vice Presidency, World Bank I would like to thank the Ministry of Finance (in particular the International Cooptation Department, International Economic and Financial Cooperation Department, Budget Department, Human Resources Development Center and the Finance Department of Guangxi Zhuang Autonomous Region for their initiative to co-host the international workshop on Subnational Debt Management and Restructuring with the World Bank. The topic of the workshop is highly relevant for China today. The proceedings of the workshop highlighted five messages:  There is a need for a transition strategy for subnational financing;  Use debt sustainability as a guide for transition;  There is no free lunch;  Strengthening subnational budget and debt management cannot wait; and  It is all about incentives. IV. There is a need for a transition strategy for subnational financing China has made remarkable economic progress in the past couple of decades: over 700 million people have been lifted out of poverty, the urban infrastructure has been developed quickly and social services have improved significantly. Investments by subnational governments have helped achieve this and sustain growth in China. However, the magnitude and the way these investments have been financed have led to a large and growing stock of subnational debt, and, if unchecked, becoming a problem of debt sustainability in a number of subnational entities. In response and to avoid a hard landing of the economy, the Government has decided to transition to a more sustainable framework for subnational financing. The 2014 Budget Law reform is a significant step in this direction. But establishing that new financing framework will not happen overnight. It will take some time to establish well-functioning institutions, policies and the legal framework that will allow subnational governments to borrow more responsibly. At the same time, spending needs are pressing and large. Among other things, the Chinese population is aging and the social sector spending pressures—in particular on health and pensions—are rising. Also, infrastructure still needs some attention. Therefore, the first message that came out clearly in the workshop is that a transition strategy from the current to a new financing framework is needed. The other messages from the workshop proceedings focused on that transition strategy. V. Use debt sustainability as a guide for transition The introduction of provincial and local government bonds—or swap of RMB3.2 trillion of existing debt to bonds--was a welcome measure. In addition to reducing debt servicing costs, it enhanced the transparency of subnational financing arrangements. 98 While that was an important step, the workshop presentations highlighted that debt sustainability analysis is needed at the subnational level to determine whether the debt swap and other measures already taken are sufficient to deal with the stock of debt. If the analysis shows the measures have not been enough, fiscal adjustment—spending and revenue measures—will be needed over time to bring debt levels under control and on a sustainable path. Further, if fiscal consolidation alone is inadequate to bring debt to sustainable levels in a reasonable time, the government will need to consider debt re- profiling or debt restructuring in highly indebted provinces. The workshop discussions highlighted that, in parallel, it is advisable to start the review of the intergovernmental fiscal framework to ensure that tax assignments match expenditure responsibility. In that context it would be important to ensure that equity concerns are addressed as reliance on market financing will tend to favor richer areas. Also, while the new financing framework can be expected to include fiscal rules, during the transition period the government could use administrative controls to set debt limits that cover all sources of financing. Further, it would be important to establish also guidelines for use of land leases to finance infrastructure. VI. There is no free lunch When preparing a transition strategy and assessing debt sustainability, international experience stresses the importance to take a consolidated “public sector” view of finances at the subnational level: including UDIC activities, estimates for contingent liabilities and off-budget expenditures. As in other countries, financing public expenditure outside the budget or through the “backdoor”, such as through extra-budgetary funds, special purpose vehicles, bank loans or PPP projects only postpones when the government has to pay. The bill always comes. There is no free lunch. VII. Strengthening subnational budget and debt management cannot wait The fourth message from the workshop was that strengthening subnational budget and debt management cannot wait. Four ways to improve subnational budget management was highlighted: (i) Integrate budgeting for capital and current expenditure and take a multi-year view; (ii) Improve public investment management to ensure that high quality projects are financed since resources are scarce; (iii) Strengthen public debt management to ensure sound borrowing decisions; (iv) Increase the use of capital market financing and PPPs. The results of these measures will take some time to show. Thus, it is important to start early. VIII. It is all about incentives None of this works without appropriate incentives to subnational governments. Peter Drucker once said that “culture eats strategy for breakfast”. To avoid that happening in China, the culture of subnational financing would benefit from some rebalancing. Specifically, at the subnational and national levels the focus should emphasize not just aiming for growth and investment, but also debt and fiscal sustainability and productive projects. The task ahead is complex, but rewarding. I look forward to further exchanges on this important agenda. 99 Xu Xiang Deputy Director-General of Human Resources Development Center, Ministry of Finance, China Thank you very much, Ms. Satu Kahkonen for your concluding remarks. I’d like to thank the technical team from the World Bank led by Ms. Satu Kahkonen for their dedication and devotion, without which this workshop could never be that high-level and professional. We also believe that this Summit, although concluded now, symbolizes starting point of further cooperation between the Ministry of Finance and the World Bank. I expect that such sound cooperation between the two parties will bear more abundant fruits in the future. Ladies and gentlemen, dear friends, China like the rest of the world has experienced the transformation from a simple government debt management practice to a more complicated one, from the theoretical study to the integration of theory and practice. I believe, through the experience, communication, and collision of ideas among Chinese and overseas experts, we now view government debt management from a wider angle and have gained deeper understanding. It will have a positive effect on the subnational debt management reform in our country. Now let us again give a warm applause to all the experts present here. Thank you for sharing with us your great knowledge. I announce that the international workshop on subnational debt management today is concluded now. Thank you. 100 Authors’ Biographies Roy Bahl is Regents Professor of Economics and Founding Dean, emeritus, Andrew Young School of Policy Studies, Georgia State University. Bahl is an expert in public finance with extensive experience consulting on state and local government issues with public and private agencies, including Standard & Poor’s Corp., several city and state governments, and federal agencies. Bahl headed the staffs of the Ohio and Georgia Tax Commissions. He has served on the editorial boards of the National Tax Journal, The Journal of Public Budgeting and Finance, and Growth and Change. He is the author of the book Fiscal Policy in China, and with Johannes Linn, of Urban Public Finance in Developing Countries. He served as principal economic advisor to IBM’s Worldwide Tax Practice. Bahl has consulted on fiscal matters with governments in developing and transition economies all over the world, as well as with the International Monetary Fund, the Asian Development Bank, the U.S. Agency for International Development, the United Nations and the World Bank. Bahl is the author of numerous books, monographs and scholarly papers in the area of urban/regional economics, public finance and economic development. He came to Georgia State from Syracuse University, where he was the director of the Metropolitan Studies Program and Maxwell Professor of Political Economy. Qiao Baoyun, Professor of Economics and Dean, China Academy of Public Finance and Policy, Central University of Finance and Economics, He published numerous books and scholarly papers. He has advised many countries on public finance, local governance, and economic development. Baoyun graduated from Georgia State University, U.S., where he earned his Ph.D. in economics. Fernando Blanco, a Bolivian national, is Lead Economist in the Macroeconomic and Fiscal Management Global Practice of the World Bank in the Latin America and Caribbean Region. He joined the World Bank in 2002 as country Economist for Brazil and became Senior Economist in 2008. Mr. Blanco worked actively in analytical and advisory projects on fiscal federalism and in the World Bank lending program in fiscal decentralization and with sub-national governments in Brazil and West Africa where he led a number of innovative projects. Mr. Blanco has worked in analytical pieces on public spending and fiscal decentralization in Brazil, Mexico, Pakistan, Kenya and Nepal. Prior to the World Bank, Mr. Blanco worked at the 101 Institute of Applied Economics Research (IPEA), a think-tank of the Brazilian government, and was professor of Public Economics and International Economics at the Brazilian Institute of Capital Markets (IBMEC) and the Pontifical Catholic University of Rio de Janeiro. Fernando holds a PhD in Economics from the Pontifical Catholic University of Rio de Janeiro with specialization in public economics and fiscal federalism theory and practice. His work on fiscal federalism issues in Brazil was awarded in 1996 with the XXI National Prize of Master Dissertations in Economics, National Bank of Economic and Social Development (BNDES) in 1996. Since 1995 is member of the Brazilian Econometric Society. Sudarshan (Shan) Gooptu, Global Lead, Fiscal Policy in the World Bank Group's Global Practice for Macroeconomics and Fiscal Management. In this role, Dr. Gooptu is the technical lead of the cross-section of staff in the Practice that have professional interests and abilities in this business line (Global Solutions Group) of the World Bank Group to help deliver strategic, global or country outcomes. He has a Ph.D. in Economics from the University of Illinois, Urbana-Champaign, U.S.A. and over 25 years’ experience in the World Bank where he has undertaken a full range of activities and operational products in both low and middle-income countries--from budget support lending operations to core economic reports, policy and technical papers, and technical assistance services. His assignments have included work on debt and fiscal risk management; public expenditure reviews, emerging markets' sovereign debt buyback and restructuring operations; policy-oriented macroeconomics research. He was the World Bank’s Lead Economist for China and Mongolia between 2001-2008, after which he served as the Sector Manager for the World Bank’s, Economic Policy and Debt Department and, until recently, a Practice Manager in the Macroeconomics & Fiscal Management Global Practice (2008 and mid-2015). Bert Hofman is Country Director, China, Mongolia and Korea. Mr. Hofman, a Dutch National, is the World Bank's Country Director for China, Mongolia and Korea in the East Asia and Pacific Region, based in Beijing. Prior to his present assignment, Mr. Hofman was the World Bank's Chief Economist for the East Asia and Pacific Region and Director, Singapore Office. As regional chief economist he led a team to analyze key trends and policy issues across East Asia and the Pacific and as Director Singapore he helped build a partnership that focuses on expanding investment in infrastructure in emerging economies. Before moving to Singapore in 2011, Mr. Hofman was the Country Director for the Philippines, responsible for a growing portfolio of projects and advisory services to the Philippines government. Hofman has accumulated more than 22 years of experience in the World Bank, 16 of which in the East Asia region. Among others, Mr. Hofman was Lead Economist for China and for Indonesia and country economist for Mongolia and Namibia. He had also worked on Brazil, South Africa, Mongolia, Zambia, and Namibia in his earlier years with the Bank. 102 Prior to joining the World Bank, Mr. Hofman worked at the Kiel Institute of World Economics in Germany, the Organization for Economic Cooperation and Development (OECD) in Paris, and NMB Bank in the Netherlands (now ING). Mr. Hofman holds a degree in economics of Erasmus University Rotterdam and studied at the Christian Albrechts University, Kiel. Satu Kahkonen is Director of the Macroeconomics and Fiscal Management Global Practice at the World Bank Group, which comprises over 350 staff with expertise in macroeconomics, fiscal policy, growth and statistics. Here, she oversees the work of the Pract... ice in Europe and Central Asia, East Asia and Pacific, as well as South Asia. Prior to this, Ms. Kahkonen was a Sector Manager in the Poverty Reduction and Economic Management unit. The cluster focused on economic policy management and covered 18 countries of the East and South East Europe—new EU Member States and Western Balkans—as well as select Eurozone countries. Preceding that, she worked as a Lead Economist in South Asia, East and South East Europe, and Africa in the World Bank. In that role, she led a team which provided economic policy advice and technical expertise to multiple countries including Pakistan, Bulgaria, Croatia, and Uganda. Before joining the World Bank, Ms. Kahkonen was part of the faculty of the University of Maryland and Associate Director of the Center for Institutional Reform and the Informal Sector (IRIS), a think tank affiliated with the Department of Economics and headed by the late Professor Mancur Olson. Ms. Kahkonen is the author of three books on economic development, and has published numerous economic reports and research papers. Ms. Kahkonen holds a Ph.D. in Economics from the University of Pennsylvania, where she studied as a Fulbright Scholar, and an M.Sc. from the Helsinki School of Economics. Lili Liu is the Global Leader on Decentralization and Intergovernmental Relations in the Governance Global Practice Group of the World Bank. She has led policy research on multilevel government finance in the context of macroeconomic adjustment, infrastructure financing, and capital market development. She co-chairs the Decentralization and Subnational Regional Economics Thematic Group, a Bankwide network with over 300 members. She has published widely including books Until Debt Do Us Part and International Experience on Subnational Debt Management. Dr Liu has been invited to speak at high-level international conferences including those organized by APEC, European Commission, the International Monetary Fund, and the governments of China, India, Italy, Poland, Russia, and Turkey. Dr. Liu has led advisory services to the World Bank’s operations in many developing countries on multilevel government finance, infrastructure financing, and capital market development. She operational work has 103 included leading high-level policy dialogue and lending operations for India and other countries, covering development strategy, macroeconomic policy, public finance, trade, and infrastructure. She has a PhD and MA in Economics from the University of Michigan, Ann Arbor. John Litwack became the Lead Economist for China as of October 2015. Before he was the Lead Economist for Nigeria from September 2010. Litwack joined the World Bank in 2002, and has been Lead Economist for Kazakhstan and the Chief Economist for Russia. Before joining the World Bank, Dr. Litwack headed the Russia Desk in the Economics Department of the Organization of Economic Cooperation and Development (OECD) from 1995-2002. 1995-1998 he was an Assistant Professor in the Economics Department at Stanford University. Dr. Litwack received his Ph.D. in economics from the University of Pennsylvania in 1998. He has published a number of books and articles on economic theory and topics of economic transition. He has also done work on the economies of Ukraine, Belarus, Bulgaria, Serbia and Montenegro. Jorge Martinez-Vazquez is Regents Professor of Economics at Georgia State University and Director of the International Center for Public Policy. He has published over 20 books and numerous articles in academic journals, such as Econometrica and Journal of Political Economy. He was the director of the $20 million USAID Fiscal Reform Project in the Russian Federation (1997-2000) and has consulted with federal agencies and state governments in the United States and over 60 countries including Russia, China, South Africa, Indonesia, and Mexico. He has consulted with the World Bank, the Asian Development Bank, the United Nations, and the Inter-American Development Bank, and is a Member of the IMF Panel of Fiscal Experts. Martinez-Vazquez has been recognized with the Russian Federation National Prize in Economics for 2007- 2009 and the 2010 Public Policy Award of the German Development Cooperation from the International Institute of Public Finance. In 1997 he was inducted into the Russian Academy of Natural Sciences and in 2006 he was made an Honorary Professor of the Central University of Finance and Economics, Beijing. He was awarded an honorary doctorate from the University of Vigo in 2011 for his extensive body of work in fiscal decentralization and taxation. Abha Prasad has over 25 years of experience working on debt management and fiscal policy. She is the Program Manager of the Debt Management Facility (DMF) and is working as a Lead Debt Specialist with the World Bank’s Macro Fiscal Global Practice. Prior to this, she was the Director with the Reserve Bank of India, the central bank, managing the debt of both the federal and 28 sub-nationals. She was actively involved in providing policy advice for framing India’s fiscal responsibility legislation. At the World Bank, as part of a core team she has been leading the work on the Debt Management Performance Assessment (DeMPA) framework, and the 104 medium term debt management strategies. She has worked in several developing countries and emerging markets on debt management issues, that include, among other, Bosnia, Brazil (Rio de Janeiro), Belarus, China (Chongqing and Hunan), Ghana, India, Indonesia (DKI Jakarta), Peru (Lima), Moldova, Nicaragua, Nigeria, Sri Lanka, Tanzania, and Vietnam. Liu Shangxi, Professor of Economics, Director of Research Institute for fiscal Science (RIFS), Ministry of Finance, People's Republic of China. Research interests are on Macroeconomics, Taxation and Public Distribution, Public Finance and Public Policy, Public Debt and Public Risk, Public Budgeting and Financial Management. He has Ph.D. in Economics from China Ren Min University. Christoph Ungerer is an Economist at the World Bank. He works in the East Asia Pacific Unit of the Macroeconomics and Fiscal Management Global Practice, EFI Vice Presidency. Previously, Christoph worked as Economist at the Office of Financial Stability, US Federal Reserve Board, and at the Country Desk for Greece, DG Economic and Financial Affairs, European Commission. Christoph holds a PhD in Economics from the LSE, an MPhil in Economics from the University of Cambridge and a BSc in Economics from the University of Nottingham. He was also a Visiting Fellow at Harvard University. Mathew Verghis is Practice Manager of Macroeconomics and Fiscal Management Department for East Asia and Pacific. Prior to assuming this position, he was Lead Economist and Sector Manager for 7 years at the World Bank for Cambodia, Lao PDR, Malaysia, Myanmar and Thailand. He had also spent fifteen years at the World Bank covering a number of countries in Europe and Africa. He has written widely on growth, poverty, fiscal policy and labor markets. He has a PhD in Economics from the University of Notre Dame in the United States. 105