86082 Generating Sustainable Wealth from Mozambique’s Natural Resource Boom 1 World Bank Mozambique - Policy Note January 2014 1 This note has been prepared by a team led by Enrique Blanco Armas, Senior Economist in the World Bank, Mozambique (AFTP1), and includes Ekaterina Gratcheva (FABBK), Dmitry Pevzner (FABRP), and Natasha Sharma (AFTP1). Assistance was provided by Peter Fisker (AFTPM). Sean Lothrop (AFTP4) edited the policy note. Comments were received by Apurva Sanghi (AFTP2), Francisco Carneiro (LCSPR), Bernand Murira (FABRP) and Alex Segura (IMF Mozambique Resident Representative). Feedback and support was provided by Julio Revilla (AFTP1) and the work was carried out under the overall guidance of John Panzer (AFTP1).The team would like to acknowledge the financing provided by DFID for this work under the programmatic AAA 'Enhancing macroeconomic and fiscal policy making for inclusive growth in a resource-rich setting'. Contents Sumário Executivo .....................................................................................................................................................3 Executive Summary ...................................................................................................................................................8 Introduction ........................................................................................................................................................12 I. MANAGING RESOURCE REVENUES ...................................................................................................................................14 A. Special Characteristics and Unique Policy Challenges ........................................................................................14 Expenditure-Capacity Constraints.......................................................................................................................14 Revenue Volatility ...............................................................................................................................................15 Finite Resources ..................................................................................................................................................16 Macroeconomic Impacts ....................................................................................................................................18 Governance .........................................................................................................................................................20 B. Developing a Strong Fiscal-Policy Framework ....................................................................................................21 Using Fiscal Targets to Achieve Policy Objectives ...............................................................................................24 The Role of Fiscal Rules .......................................................................................................................................25 Improving the Government’s Capacity to Manage Public Spending ..................................................................26 Optimizing the impact of Resource Wealth through Increased Public Savings ..................................................31 Establishing a National Development Bank ........................................................................................................36 Managing Assets and Liabilities ..........................................................................................................................38 The Importance of Public Expectations and Support for Resource-Revenue Policies ........................................38 II. DESIGNING A FISCAL-POLICY FRAMEWORK TO MANAGE NATURAL-RESOURCE REVENUES IN MOZAMBIQUE ...............................41 The Estimated Fiscal Envelope from Both Resource and Non-Resource Revenues ...........................................41 The Impact of Fiscal Rules ...................................................................................................................................44 Projections for Public Savings .............................................................................................................................46 III. CONCLUSION AND RECOMMENDATIONS..........................................................................................................................47 Annex 1: Suggestions for how the Medium Term Fiscal Framework (MTFF) can be improved ...........................51 Annex 2: Key Tenets of the Santiago Principles for Managing Sovereign Wealth Funds .....................................52 Annex 3: Statistics for Scenarios ............................................................................................................................53 References ..............................................................................................................................................................54 2 Sumário Executivo O desenvolvimento dos sectores de gás e carvão em Moçambique representa uma oportunidade ímpar para acelerar o crescimento inclusivo e reduzir a pobreza. Caso os progressos nos sectores do gás e do carvão ocorram conforme planificado as receitas públicas irão sofrer um aumento dramático; as receitas provenientes dos recursos naturais podem atingir os 9 biliões de dólares americanos em 2032, representando 7 por cento do Produto Interno Bruto (PIB) e 21 por cento das receitas totais do Estado. O enorme crescimento que se espera da indústria extractiva Moçambicana pode reduzir drasticamente a pobreza a médio prazo e ajudar a criar as bases para um crescimento sustentável e uma prosperidade partilhada. Do mesmo modo que acontece noutros países ricos em recursos naturais, a gestão dos recursos naturais e a maximização dos seus benefícios constituem um grande desafio para Moçambique. A utilização efectiva da riqueza proveniente dos recursos naturais coloca desafios complexos para a política macroeconómica e fiscal. O rápido crescimento das receitas públicas colectadas pode ultrapassar a capacidade de absorção das entidades governamentais. Os preços das mercadorias são extremamente voláteis. Esta volatilidade poderia ser transferida para o orçamento excepto se forem tomadas medidas de estabilização. O crescimento excessivo das despesas públicas pode também afectar a estabilidade macroeconómica a curto prazo. A exportação de recursos naturais de valor elevado pode também causar a apreciação da taxa de câmbio real afectando a competitividade das indústrias que não dependem dos recursos naturais. Moçambique deve aprovar um quadro de política fiscal que tome em consideração as características das receitas dos recursos naturais. Os indicadores tradicionais da política fiscal não serão adequados para avaliar a posição fiscal do país devido as características das receitas dos recursos naturais. Os países ricos em recursos naturais tem estado cada vez mais a aprovar quadros de política fiscal que os ajudam a atingir os seus objectivos de política fiscal ao mesmo tempo abordando os desafios acima debatidos. O quadro de política fiscal vai ajudar a tomar decisões sobre o nível de gasto público e poupanças, a manutenção da estabilidade macroeconómica a curto prazo e a não sobrecarregar a capacidade de gestão do governo bem como proteger o orçamento contra a volatilidade de preços das mercadorias. Moçambique deve aprovar o Quadro de Sustentabilidade Fiscal (Fiscal Sustainability Framework – FSF), um quadro de política fiscal que permite o aumento dos investimentos públicos com o objectivo de acelerar o crescimento nos sectores que não dependam de recursos naturais. O FSF permite a realização antecipada de despesas públicas com vista a satisfazer as necessidades de desenvolvimento de Moçambique. A antecipação das despesas públicas deverá gerar níveis mais elevados de retorno através do crescimento acelerado ao invés de apenas proceder a acumulação de poupanças. O aumento em termos de gasto público deverá ser ajustado à capacidade de absorção da economia e à capacidade do governo de gerir um leque mais amplo de recursos naturais. As regras fiscais iriam determinar quanto pode ser gasto e quanto deve ser poupado. As receitas que ultrapassem a capacidade de absorção do governo deverão ser depositadas num fundo soberano de riqueza nacional. As poupanças seriam usadas para efeitos de estabilização e para serem gastos no futuro. Para que funcione devidamente, as regras e metas fiscais devem estar ajustadas as circunstâncias específicas de 3 Moçambique, e devem ser alvo de apoio político amplo e de um compromisso credível do governo. A implementação de tal quadro fiscal será complexa e exige a capacitação do Governo em várias áreas. A dimensão na qual o FSF vai atingir um crescimento mais elevado vai depender da qualidade das despesas públicas. Um nível de gasto público maior, se não for bem gasto, pode não conseguir gerar os retornos esperados. Neste contexto, os progressos contínuos na reforma da Gestão das Finanças Públicas (GFP) afiguram-se fundamentais para as expectativas de longo prazo de Moçambique. Em primeiro lugar, as receitas dos recursos naturais devem ser incluídas em toda a documentação orçamental, em especial o CFMP. Em segundo lugar, é fundamental criar projecções precisas e credíveis de receitas para a planificação multianual de investimentos. Em terceiro lugar, os processos de execução e controlo orçamental devem ser fortalecidos de forma que a integridade dos sistemas de GFP não sejam comprometidos a medida que as receitas aumentam. Em quarto lugar, é importante reforçar a abrangência e transparência da GFP mediante a prestação de contas de todas as operações governamentais, incluindo gasto público extra-orçamental e os passivos contingentes enquanto continuando a fortalecer o acesso público à informação fiscal chave. Finalmente, a melhoria da responsabilização e prestação de contas externas através de auditorias regulares e supervisão pela AR vai fortalecer a integridade do sistema de GFP. Quando o governo tiver reforçado de forma adequada a GFP o mesmo deve começar a mudar para um cenário de despesas a médio prazo (CDMP), um instrumento de política que oferece vantagens consideráveis para os países ricos em recursos naturais. O CDMP providencia aos fazedores de políticas uma base para a gestão de riscos e das incertezas numa perspectiva a médio prazo. Isto é particularmente importante para a mitigação da volatilidade e o gasto público pro-cíclico associados as receitas dos recursos naturais. A abordagem de médio prazo de planificação e orçamentação para projectos de investimento e em particular a inclusão dos custos de Operação e Manutenção ( O&M) nos orçamentos, vão ajudar a reforçar a sustentabilidade fiscal a medida que os níveis de investimento aumentam. Mediante a inclusão dos riscos fiscais, a sustentabilidade da dívida e dos passivos contingentes o CDMP promove uma boa abordagem inter-temporal da política fiscal. Moçambique precisa de melhorar o seu sistema de gestão dos investimentos públicos (GIP) conforme o governo continua a aumentar o investimento público. Um sistema forte de GIP será um elemento fundamental para garantir que os gastos gerem o retorno esperado. A actual capacidade de Moçambique na GIP é relativamente fraca. A classificação de Moçambique é particularmente baixa no que respeita a avaliação de projectos. Isto é um elemento importante da GIP visto que o processo de avaliação é fundamental para a selecção dos projectos com elevados retornos económicos e sociais. Estão a ser envidados esforços no sentido de melhorar o processo através do qual os projectos de investimentos públicos são avaliados, analisados e seleccionados em Moçambique. Os esforços planificados e em curso irão, sem dúvidas, melhorar a capacidade do Governo de gerir investimentos públicos, mas isto leva tempo e passaram vários anos antes de Moçambique ter um sistema robusto de GIP e do devido fortalecimento das capacidades do sector público. Qualquer aumento planificado nos investimentos públicos deve tomar em consideração estes constrangimentos de capacidade o que justificaria uma abordagem mais gradual no incremento dos investimentos públicos. 4 O quadro de política fiscal sugerido, o FSF, vai resultar em poupanças significativas a partir das receitas dos recursos naturais. Moçambique vai precisar de uma estratégia para a gestão destas poupanças. As receitas que ultrapassam a capacidade de absorção do país devem ser poupadas num Fundo de Riquezas Soberano (FRS). Tal fundo pode adiantar uma série de objectivos importantes, protegendo o orçamento contra a volatilidade das receitas, mantendo as despesas públicas de acordo com a capacidade das entidades públicas, mitigando o impacto macroeconómico do aumento das despesas e a promoção da equidade intergeracional. As poupanças podem ser mantidas num fundo com dois principais objectivos. Um fundo de estabilização resolve os problemas relacionados com a volatilidade das receitas dos recursos naturais. Quando os preços das mercadorias estiverem elevados as receitas excedentes são depositadas no fundo e quando os preços baixam são feitos levantamentos para cobrir o défice orçamental resultante. Os fundos de estabilização devem estar disponíveis para levantamento atempado e os investimentos devem focalizar em instrumentos de baixo risco e liquidez elevada. Os fundos de poupança preservam a riqueza para gerações futuras e cria activos financeiros para compensar a diminuição da riqueza dos recursos naturais. Os fundos de poupança possuem horizontes de investimento mais longos e envolvem investimentos de maior risco e menor liquidez. As duas funções não são necessariamente incompatíveis e em muitos países o mesmo FRS serve para ambos os efeitos. O FRS deve estar situado dentro de um quadro institucional forte e estreitamente ligado ao ciclo orçamental. O fundo deve ser integrado com o orçamento e os levantamentos devem ser feitos como parte de um processo unificado de execução orçamental. O governo deve ser o proprietário do fundo com o Ministério das Finanças a agir em sua representação. O parlamento aprova as leis que criam a sua estrutura formal e as normas para a sua gestão. O Banco Central deve gerir o fundo, visto que o mesmo possui capacidade técnica para gerir activos financeiros em consonância com as suas funções de política cambial e monetária. O governo deve criar uma divisão clara e transparente das funções e responsabilidades entre as instituições envolvidas com o FRS, particularmente o proprietário do fundo e o gestor. Este aspecto é fundamental para minimizar a interferência política e promover a gestão dos activos baseada exclusivamente em critérios técnicos. Com vista a evitar influência política indevida na gestão do fundo é necessário limitar o poder discricionário dos fazedores de políticas através de normas simples, claras e exequíveis (por exemplo, para depósitos e levantamentos) que evitem gastos excessivos e salvaguardem a integridade da gestão do FRS. Será necessário tomar decisões sobre onde investir a poupança nacional. Muitos FRS investem exclusivamente em activos estrangeiros, mas existe um interesse crescente em muitos países ricos em recursos naturais no investimento em activos domésticos. Os investimentos domésticos podem ser usados para promover a diversificação económica e evitar os efeitos da Doença Holandesa (‘Dutch Disease’) elevando a competitividade dos sectores que não dependam de recursos naturais tais como a indústria manufactureira. Nalguns casos, a taxa de retorno dos investimentos internos pode ser maior do que os investimentos no estrangeiro. Mas existem uma série de riscos relacionados com uma carteira de investimentos focalizada em investimentos domésticos. Uma estratégia para a diversificação dos riscos iria sugerir tipicamente o investimento em activos no estrangeiro uma vez que o valor dos activos domésticos (e por conseguinte a capacidade do governo de efectuar levantamentos do fundo) iriam 5 diminuir perante uma desaceleração da economia domestica. A busca de retornos elevados pode levar o fundo a investir em áreas que já estão a receber investimentos privados significativos (numa economia que pode estar em sobreaquecimento). As decisões sobre onde investir devem tomar em conta os objectivos de Moçambique bem como os riscos envolvidos e alcançar uma carteira de investimentos equilibrada. O Governo de Moçambique manifestou a sua intenção de criar um banco de desenvolvimento com vista a preencher algumas das lacunas no sector financeiro do país, mas este plano deve ser considerado com precaução devido aos significativos riscos envolvidos. O principal motivo para a utilização das receitas dos recursos naturais para capitalização de um banco de desenvolvimento é para ultrapassar os constrangimentos no mercado financeiro. O mandato de um banco de desenvolvimento é normalmente o de conceder crédito e serviços financeiros aos clientes que não são cobertos pelo sistema financeiro existente. Isto significa que as Pequenas e Medias Empresas (PMEs) são normalmente as maiores beneficiárias de um banco de desenvolvimento. Em muitos países, os bancos de desenvolvimento faliram, gerando instabilidade financeira e muitas vezes forcando o governo a intervir a um custo fiscal considerável. As lições aprendidas da experiência internacional e da própria história de Moçambique destacam riscos significativos associados à constituição de um banco de desenvolvimento. A experiência internacional tem produzido uma série de princípios de boas práticas que Moçambique deve seguir caso constitua um banco de desenvolvimento. O banco deve possuir um mandato claro, incluindo um sector alvo bem identificado. Deve possuir uma boa estrutura de controlo institucional, incluindo um conselho de administração profissional e os gestores devem ser capazes de administrar livres de interferência política. Deve ser prestada atenção especial aos procedimentos de gestão do risco e a percentagem de crédito mal parado no balanço do banco. Com vista a garantir que o banco de desenvolvimento materialize os seus objectivos ao mesmo tempo minimizando riscos tanto ao sector financeiro bem como a posição fiscal do Estado é fundamental a existência de regulamentos fortes e de uma supervisão activa. Um banco de desenvolvimento deve focalizar nas lacunas do mercado financeiro, incluindo a falta de financiamento a longo prazo e o acesso ao crédito por parte das PMEs, complementando as instituições financeiras existentes e não concorrendo com elas. Um ambiente forte de controlo institucional é fundamental para o sucesso de banco de desenvolvimento e sem esse ambiente os riscos são tão significativos que seria preferível Moçambique tentar ultrapassar os constrangimentos no mercado financeiro de outras formas. Com vista a coordenar as suas cada vez mais sofisticadas operações financeiras o Governo de Moçambique deve analisar a possibilidade de adoptar uma política mais ampla de gestão de passivos e activos soberanos. As receitas provenientes da indústria extractiva criam o seu próprio conjunto de desafios e oportunidades para a gestão dos activos e passivos soberanos devido ao acesso a novos instrumentos financeiros e opções de financiamento com divida. Uma política de gestão de activos e passivos deve incluir tanto os activos como os passivos (implícitos e explícitos) públicos incluindo os programas sociais, obrigações resultantes das empresas públicas, garantias prestadas tanto às empresas públicas bem como privadas, fundos de pensões, etc. A formulação de objectivos viáveis vai exigir a avaliação sistemática da sustentabilidade fiscal, análise da sensibilidade das variáveis que influenciam o 6 ciclo macroeconómico e a análise da estrutura e da exposição ao risco do balanço do governo. Sem um quadro abrangente, a acumulação de passivos contingentes e das despesas fora do orçamento poderia comprometer uma posição fiscal que parece equilibrada. A apresentação de toda a informação relativa aos activos e passivos públicos de uma maneira clara e abrangente seria o primeiro passo no sentido de melhoramento da transparência e a prestação de contas e duma abordagem mais abrangente dos desafios de um sistema financeiro público mais complexo. A mitigação dos riscos políticos para as politicas responsáveis de gestão dos recursos naturais exigem a gestão das expectativas dos cidadãos e a criação de um apoio doméstico sustentando à estratégia do governo para a gestão dos recursos naturais. A insatisfação popular com o rumo do crescimento das despesas públicas ou a alocação destas pode provocar um conflito social, cuja probabilidade é mais elevada num país pós-conflito como é o caso de Moçambique. As discussões públicas devem focalizar na promoção de expectativas realistas relativas à quantidade e ao prazo das receitas dos recursos naturais. Processos consultivos e participativos mais amplos para determinar os objectivos do quadro da política fiscal irão aumentar a apropriação pública das decisões sobre as despesas e poupanças, e se devem organizar consultas frequentes com vista a identificar prioridades para as despesas públicas. A sustentação da confiança pública na gestão dos recursos naturais requer que se reforce de forma continua a transparência e prestação de contas. O governo deve publicar com regularidade toda a informação pertinente sobre as receitas dos recursos naturais nos seus documentos de orçamento, incluindo o “Orçamento Cidadão”. Outras medidas para melhorar a prestação de contas externas, em particular a supervisão da AR e auditoria externa bem como o cumprimento das novas normas da Iniciativa de Transparência na Industria Extractiva (ITIE) poderiam ajudar a reforçar a confiança pública. Os requisitos de cumprimento da ITIE e outras medidas complementares não obrigatórias mas importantes incluiriam a disponibilização de informação contextual sobre as indústrias extractivas, o alargamento do âmbito da prestação sectorial de contas mediante a divulgação completa dos termos de todos os contratos de extracção de recursos naturais e a publicação de informação sobre todas as adjudicações de novas licenças e concessões. 7 Executive Summary The development of the coal and gas sectors in Mozambique represent an unprecedented opportunity to accelerate inclusive growth and reduce poverty. If the developments in the coal and gas sectors proceed as planned public revenues will increase dramatically; resource revenues could be as high as US$9 billion by 2032, representing 7 percent of GDP and 21 percent of total government revenues. The coming boom in Mozambique’s extractive industries could dramatically reduce poverty in the medium term and help to establish the foundation for sustainable growth and shared prosperity. As in many other resource-rich countries, it will be challenging to manage natural resources and maximize the benefits for Mozambique. Effectively utilizing resource wealth poses complex challenges for macroeconomic and fiscal policy. Rapidly rising public revenues may outstrip the absorptive capacity of government agencies. Commodity prices are very volatile. This volatility is often transferred to the budget unless stabilization measures are taken. Excessive public expenditure growth may also affect short-term macroeconomic stability. High-value resource exports may also cause an appreciation of the real exchange rate, damaging the price competitiveness of non-resource industries. Mozambique should adopt a fiscal policy framework that takes the characteristics of natural resource revenues into account. Traditional fiscal policy indicators will not be appropriate to assess the country’s fiscal stance given the characteristics of natural resource revenues. Resource-rich countries have increasingly adopted fiscal frameworks that help them achieve their fiscal policy objectives while addressing the challenges discussed above. The fiscal framework will help make decisions about how much to spend and save, maintain short term macroeconomic stability and not overwhelm the government’s management capacity, as well as insulate the budget from commodity price volatility. Mozambique should adopt the Fiscal Sustainability Framework (FSF), a fiscal policy framework which allows for increasing public investments with the aim of accelerating growth in the non-resource sector. The FSF would allow for a frontloading of public spending to meet Mozambique’s development needs. Frontloading spending should generate higher returns through accelerated growth, instead of accumulation of savings alone. The increase in public spending should be in line with the economy’s absorptive capacity and the government’s capability to manage a larger resource envelope. Fiscal rules would determine how much can be spent and how much needs to be saved. Revenues that exceed the government and the economy’s absorptive capacity would be saved in a sovereign wealth fund. The savings would be used for stabilization purposes and to be spent in the future. In order to function properly fiscal targets and rules must be tailored to the specific circumstances of Mozambique, and they must enjoy broad political support and a credible government commitment. Implementation of such a framework will be complex and require building the Government’s capacity in a number of areas. The extent to which the FSF achieves higher growth will depend on the quality of public spending. Increased public resources, if not well spent, could fail to generate the expected returns. In this context, continued progress in PFM reforms is key to Mozambique’s long-term prospects. First, resource revenues must be included in all budget documentation, particularly the MTFF. Second, it is essential to establish accurate revenue projections and credible multi-year investment planning. Third, budgetary 8 controls and budget-execution processes must be strengthened so that the integrity of public financial management systems is not compromised as revenue levels increase. Fourth, it is important to reinforce the comprehensiveness and transparency of public financial management by reporting on all government operations, including extra-budgetary operations and contingent liabilities while continuing to enhance public access to key fiscal information. Finally, improving external accountability through regular audits and legislative oversight will bolster the integrity of the PFM system. Once the government has adequately reinforced its PFM systems it should begin to shift toward a medium-term expenditure framework (MTEF), a policy tool which offers considerable advantages for resource-rich countries. An MTEF provides policymakers with a basis for managing risks and uncertainties from a medium-term perspective. This is particularly important for mitigating the volatility and procyclical spending bias associated with resource revenues. A medium-term approach to planning and budgeting for investment projects, and in particular the inclusion of O&M costs in budgets, will help to reinforce fiscal sustainability as investment levels increase. By encompassing fiscal risks, debt sustainability and contingent liabilities MTEFs promote a sound inter-temporal approach of fiscal policy. Mozambique needs to improve its public investment management (PIM) system as the government continues to increase public investment. A strong PIM system will be a crucial element in ensuring that increased spending delivers the expected results. Mozambique’s current capacity on managing public investments is relatively weak. Mozambique scores particularly low on project appraisal, which is an important element of PIM as appraisal processes are critical for selecting projects with high economic and social returns. Efforts are underway to improve the process by which public investment projects are appraised, evaluated and selected in Mozambique. The ongoing and planned efforts will no doubt improve the Government’s capacity to manage public investments, but it will be years before a robust PIM system is in place and public-sector capacities have been adequately strengthened. Any planned increases in public investments must take these capacity constraints into account and a more gradual approach may be warranted given the weaknesses discussed. The fiscal policy framework suggested (FSF) will result in significant savings from natural resource revenues. Mozambique will need a strategy to manage these savings. Revenues that exceed the country’s absorptive capacity should be saved in a SWF. Such a fund can advance a number of critical objectives, by serving as a buffer against revenue volatility, keeping public spending in line with the capacity of public agencies, mitigating the macroeconomic impact of expenditure increases and promoting intergenerational equity. Savings can be kept in a fund with two main purposes. A stabilization fund addresses the volatility of resource revenues. When commodity prices are high excess revenues are deposited into the fund, and when prices are low withdrawals are made to cover the resulting budget shortfall. Stabilization revenues must be available for withdrawal on short notice and investments should focus on low-risk, high-liquidity instruments. A savings fund preserves wealth for future generations and creates financial assets to compensate for the depletion of natural wealth. Savings funds have much longer investment horizons and involve riskier, less liquid assets. The two functions are not necessarily inconsistent and in many countries the same resource fund serves both purposes. 9 SWF must be situated within a strong institutional framework and closely linked with the budget cycle. The fund should be integrated with the budget and withdrawals should be spent as part of a unified budget-execution process. The government should be the owner of the fund with the Minister of Finance acting on its behalf. The parliament adopts the laws that establish its formal structure and the rules for its management. The central bank should manage the fund, as it has the technical capacity to manage financial assets in concert with their monetary and exchange rate functions. The government must establish a clear and transparent division of roles and responsibilities between the funds governing bodies, particularly the owner of the fund and the manager. This is essential to minimize political interference and promote asset management based solely on technical expertise. To prevent undue political influence in the management of the fund it is necessary to limit the discretionary authority of policymakers through simple, straightforward and enforceable rules (e.g. for deposits and withdrawals) that avoid overspending and safeguard the integrity of resource-fund management. Decisions will need to be made regarding where to invest national savings. Many SWFs invest exclusively in foreign assets, but there is growing interest in many resource-rich countries in investing in domestic assets. Domestic investments could be used to promote economic diversification and offset Dutch-disease effects by boosting the competitiveness of non-resource tradable sectors such as manufacturing. In some cases the rate of return on domestic investments may be higher than investment in foreign assets. But there are a number of risks associated with an investment portfolio focused on domestic investments. A strategy to diversify risks in the investment portfolio would typically suggest investing abroad, since the value of domestic assets (and therefore the government’s ability to make withdrawals from the fund) would decline in the face of a slowdown in the domestic economy. The search for high returns domestically may lead the fund to invest in areas that are already receiving significant private investment (in an economy that may be overheating). Decisions on where to invest should take both the policy objectives of Mozambique as well as the risks entailed into account and achieve a balanced investment portfolio. The Government of Mozambique has indicated its intention to establish a development bank in order to fill some of the gaps in the country’s financial sector, but this plan should be regarded with caution given the substantial risks involved. The main rationale for using resource revenues to capitalize a development bank is to address financial market failures. A development bank’s mandate is often to provide credit and financial services to clients that are not served by the existing financial sector. This means that SMEs are often major beneficiaries of a development bank. In many countries development banks have failed, generating financial instability and often forcing the government to intervene at a considerable fiscal cost. Lessons learned from international experience and from Mozambique’s own history highlight the substantial risks associated with the creation of a development bank. The international experience has yielded a number of best-practice principles that Mozambique should follow if it creates a development bank. The bank should have a clear policy mandate, including a well-identified target sector. It should have a sound corporate-governance structure, including an independent and professional board of directors, and managers must be able to operate free from political interference. Careful attention must be paid to risk-management procedures and the share of 10 NPLs on the bank’s balance sheet. Strong regulations and active supervision are essential to ensuring that the development bank advances its policy goals while posing minimal risks to either the financial sector or the government’s fiscal position. A development bank should focus on gaps in the financial market, including lack of long-term financing and access to credit by SMEs, complementing existing financial institutions and not competing with them. A strong governance environment is crucial for the success of a development bank, and without that environment the risks are so substantial that Mozambique may be better off attempting to address financial market failures through other means. To coordinate its increasingly sophisticated financial operations the Government of Mozambique should consider adopting a comprehensive sovereign asset and liability management policy. Revenues from extractive industries create their own set of challenges and opportunities for asset and liability management by affording access to new financial instruments and debt-financing options. An asset and liability policy should include both public assets and the (implicit and explicit) liabilities of the government, including social programs, obligations arising from SOEs, guarantees provided to both public and private companies, pension funds, etc. Formulating feasible objectives will require systematic assessment of fiscal sustainability, sensitivity analysis of the variables that influence the macroeconomic cycle and analysis of the structure and risk exposure of the government’s balance sheet. Without a comprehensive framework the accumulation of contingent liabilities and off-budget expenditures could jeopardize an otherwise balanced fiscal stance. Presenting all asset and liability information in a clear and comprehensive manner would be a first step towards enhancing transparency and accountability and addressing the challenges of a more complex public financial system. Mitigating political risks to responsible resource management policies requires managing the expectations of citizens and building sustained domestic support for the government’s strategy to manage natural resources. Popular dissatisfaction with the pace of expenditure growth or the allocation of expenditures could provoke social unrest, the likelihood of which is heightened in a post-conflict country such as Mozambique. Public outreach should focus on fostering realistic expectations regarding the quantity and timing of resource revenues. Broad participatory and consultative processes for determining the objectives of the fiscal policy framework will increase public ownership of spending and savings decisions, and ongoing consultation should be held to identify priorities for public spending. Sustaining public confidence in natural-resource management requires continuously reinforcing transparency and accountability. The government should regularly publish all pertinent information on resource revenues in its budget documents, including the “citizen’s budget”. Further measures to improve external accountability, particularly legislative oversight and external audit, as well as achieving compliance with the new Extractive Industries Transparency Initiative (EITI) standard, would help to reinforce public confidence. EITI compliance requirements and important non-mandatory complimentary measures would include providing contextual information on the extractive industries, expanding the scope of sectoral reporting by mandating full disclosure of the terms of all resource- extraction contracts, and publishing information on all awards of new licenses and concessions. 11 Introduction Mozambique has enjoyed strong and sustained growth since the end of its civil war in 1992. Mozambique’s economy grew at an average rate 7.4 percent over the past two decades, due in large part to sound macroeconomic management, large-scale foreign-investment projects and donor support. Mozambique’s reliance on external financing, although still significant, is quickly diminishing as improvements in tax administration boost domestic revenues. About 70 percent of the population live and work in rural areas, primarily in agriculture or related sectors, comparable to other countries in the region. Mozambique has ample arable land, water and energy resources, and its strategic location offers considerable opportunities for regional trade. Mozambique borders six other countries, including regional economic leader South Africa, and four of its neighbors are landlocked and depend at least in part on Mozambique for access to global markets. Despite these advantages, the pace of poverty reduction has slowed significantly over the past decade, and the country faces considerable development challenges across a wide range of policy areas. Robust, broad-based growth and rapidly falling poverty rates in the aftermath of the civil war, have gradually given way to less equitable growth, persistent poverty and rising inequality, with over 50 percent of the population still living below the official poverty line as of 2009. Despite improvements in some social-development indicators nonmonetary poverty remains pervasive and severe, with Mozambique ranking 185th out of the 187 countries included in the latest Human Development Index. Good governance remains a key challenge, and progress on public-sector reform has been mixed. Mozambique has made significant strides in improving public financial management (PFM), which is demonstrated in recent surveys such as the Public Expenditure and Financial Accountability (PEFA) assessment. Progress in other governance areas, however, has been slow and uneven, giving rise to concerns about the scope and pace of reforms. Mozambique ranks in the bottom 50 percent on most dimensions of governance covered by the World Bank’s Worldwide Governance Indicators.2 Mozambique has done relatively well in other governance aspects, such as political openness, free and democratic elections, a relatively free press and a politically vocal civil society. The municipal elections in late 2013, boycotted by RENAMO, as well as an escalation in violent clashes between armed members of RENAMO and government security forces, have highlighted the need to continue improving in all governance aspects. In this context Mozambique’s recent large-scale discoveries of extractable resources, including huge reserves of natural gas and substantial deposits of coal and rare-earth minerals, present the country with an unprecedented opportunity to accelerate inclusive growth and reduce poverty. If the developments in the coal and gas sectors proceed as planned public revenues will increase dramatically; resource revenues could be as high as US$9 billion by 2032, representing 7 percent of GDP and 21 2 The Worldwide Governance Indicators measure a country’s public-sector effectiveness, regulatory quality, rule of law, control of corruption, political stability and voice and accountability. 12 percent of total government revenues.3 The coming boom in Mozambique’s extractive industries could dramatically reduce poverty in the medium term and help to establish the foundation for sustainable growth and shared prosperity. However, these positive outcomes are far from certain, and achieving them will require a combination of highly responsible, prudent governance and continuous improvements in administrative capacity. Resource endowments are not an unqualified good, and the international experience has repeatedly demonstrated that extractable resources will only realize their development potential if they can be efficiently managed and effectively transformed into productive physical and human capital. To offset the decline in natural wealth caused by the extraction of coal and gas, resource revenues must be invested in areas that generate lasting improvements in productivity and social wellbeing. This includes building human capital by enhancing public health and education and expanding physical capital through investment in infrastructure. Resource revenues have unique characteristics that can make them difficult to manage. Changing commodity prices and exogenous production variables can make resource revenues highly volatile and unpredictable. Mozambique’s resource revenues are projected to fluctuate substantially. Resource revenues will increase more rapidly than the public sector’s capacity to spend them. Resource revenues are based on the extraction of a finite stock of wealth, yet governments and the general public sometimes treat that stock as though it were infinite. Any fiscal framework built upon the expectation of indefinite revenues will become unsustainable. Excessive increases in public expenditure can also affect macroeconomic stability and lead to Dutch Disease. In addition, by reducing the need for taxation resource revenues can weaken the link between the government and its people. Without broad taxation citizens have less incentive to hold their government accountable for the quality of its policies and services. Given Mozambique’s significant development needs, the Government of Mozambique should use the natural resource revenues primarily to invest in human and physical capital – bearing in mind the challenge in managing natural resource revenues. To maximize the benefits from natural resource revenues, Mozambique should increase investments in human and physical capital, which could generate significant returns in the form of higher growth. But increases in public expenditure should take into account the unique challenges of natural resources described above. A fiscal framework that takes into account these particular challenges while allowing Mozambique to frontload investments to maximize benefits would be best suited for the country. Such a framework would also generate savings that will need to be managed. The Government of Mozambique will need to implement reforms both to spend public resources better as well as manage a larger pool of savings through a sovereign wealth fund. The quality of public spending will be a key constraint on increasing public spending. Enhancing the impact of resource 3 This revenue share is comparable to that Chile or Mongolia, but far from countries that depend from commodities for a majority of their revenue. Angola, for example, derives 78 percent of its revenue from commodities, and in many oil-rich states in the Persian Gulf commodity revenues account for 70-90 percent of total revenues. 13 revenues and bolstering the integrity of the public administration require a sustained commitment to strengthening PFM systems. As the government progressively strengthens its PFM systems policymakers must determine how to manage excess revenues that outstrip the current spending capacity of public agencies. A large number of countries, in particular those rich in natural resources, have tried to manage excess revenues establishing a sovereign wealth fund (SWF), effectively a national savings account in which revenues can be vouchsafed for future use. The extent to which resource revenue management is successful will depend in large part on the quality of the institutional environment. A common feature among countries that have managed resource revenues effectively is sustained investment in the quality of public institutions. While Mozambique has made progress in strengthening its institutional environment—including revising the legal framework and fiscal regime governing the resource sector, maintaining compliance with the evolving Extractive Industries Transparency Initiative (EITI) standard4 and demonstrating a broad commitment to PFM reform—much more remains to be done. This policy note focuses on how the fiscal-management institutions and processes can be strengthened to prepare for the inflow of resource revenues. This policy note aims to contribute to a national debate on the management of natural resource revenues. This policy note, part of a broader analytical program that seeks to inform macroeconomic and fiscal policy making in Mozambique to maximize the benefits from natural resource extraction, presents a series of recommendations on managing natural resource revenues. It assumes that the Government of Mozambique will chose to frontload public investments and spend a large share of the revenue from natural resources to meet the country’s vast development needs. The policy note is divided in three sections. Section 1 discusses some of the challenges involved in responsibly managing resource revenues and key factors to consider in developing the fiscal policy framework. Section 2 presents a simulation exercise to illustrate how a fiscal policy framework that frontloads public spending can be implemented in Mozambique. Section 3 concludes the discussion and summarizes recommendations for enhancing fiscal and resource management in Mozambique. I. MANAGING RESOURCE REVENUES A. Special Characteristics and Unique Policy Challenges Expenditure-Capacity Constraints Mozambique has a relatively large public sector, and government agencies may have difficulty absorbing large increases in budget resources without compromising the quality of public spending. Public expenditures are projected to rise above 40 percent of GDP in 2014, a very high rate by the 4 The EITI provides countries and natural-resource firms with reporting guidelines for all payments and transfers between firms and the government. 14 standards of comparable countries (see Figure 1 and Figure 2 below). Given Mozambique’s administrative and institutional constraints further fiscal expansion may challenge the ability of implementing agencies to use resources effectively. The pace of public expenditure growth should be kept closely in line with a realistic appraisal of the government’s capacity to manage a growing resource envelope. Financial controls and safeguards are exceptionally important in resource-rich countries and provide an essential complement to well-paced spending increases. Recent analytical work by the IMF (Dabla- Norris et al., 2011) argues that “resource revenue windfalls frequently increase incentives to misappropriate funds, thus discouraging incentives for sound institutional and public investment management processes.” In a context of strong institutions and effective oversight moderate budgetary increases will help to create a virtuous cycle in which additional resources can be used to strengthen governance and promote more efficient public spending. Conversely, increasing budget allocations in a weak institutional environment with limited oversight will have the opposite effect, further undermining governance and encouraging corruption and waste. Figure 1: Both current and capital expenditures in Figure 2: …resulting in a very large public sector Mozambique have steadily increased… relative to comparable countries. Source: 2006-10, CGE. 2011, ROE. 2012-13, OE Source: WDI and WB staff estimates Revenue Volatility The inherent volatility of resource revenues also poses serious fiscal policy challenges. Resource revenues tend to be highly unpredictable in terms of both their price and production dimensions. Resource-export prices are tied to international commodity markets, which are prone to fluctuate based on a complex array of global supply- and demand-side variables. Meanwhile, domestic production may be subject to shocks caused by unforeseen technical problems, infrastructure bottlenecks, regional or national instability, divergence of actual resource quantities from initial estimates, among many other factors. A downturn in production in a context of low world prices can dramatically slash resource 15 revenues, while a combination of high prices and robust production can cause massive revenue spikes. The extent to which volatility is transmitted to the budget depends on both the share of resource revenues in total public revenues and the presence or absence of effective mechanisms to insulate the budget from fluctuations in resource revenues. Revenue volatility can cause serious fiscal and budgetary problems by making it difficult to project revenue levels in advance and plan expenditures accordingly. When actual revenues exceed projections policymakers often face strong pressures to allocate them for immediate spending. Shortfalls in projected revenue disrupt multi-year projects and programs, and the need to protect priority spending can lead to unsustainable fiscal deficits. In Angola in the early 2000s high commodity prices prompted an expansionary fiscal policy. When oil and diamond export prices declined in 2008 GDP contracted sharply. The economy entered a crisis, which resulted in a Stand-by Arrangement with the IMF. In Mexico from 1973 to 1981 following a short oil windfall the authorities overestimated oil prices and borrowed against future oil earnings. A combination of inappropriate policies and an unanticipated downturn in oil prices led to the 1982 Mexican debt crisis (Everhart and Duval-Hernandez, 2001). Even when revenue fluctuations are estimated accurately over the medium-term, tying budget allocations to revenue projections can generate a procyclical bias in fiscal policy, as rising resource revenues are frequently associated with periods of broader economic growth, while falling resource revenues are often either a cause or consequence of economic downturns. In addition to volatility, there is some uncertainty as to the future path of commodity prices and how the external environment could affect the development of the gas and coal sectors in Mozambique. Most analysts project significant demand growth for natural gas globally over the next 25 years, but there is considerable uncertainty regarding the path of prices for gas and coal globally. Prices for LGN will be determined primarily by demand in new markets (e.g. China or India) as well as alternative energy sources (e.g. shale gas) that may compete with LNG. A prolonged slowdown in key markets or availability of alternative energy sources in these countries could affect global LNG markets and prices. Demand for coal is partly affected by the regulatory environment and preferences for cleaner energy sources. Price projections for coal and LNG over the next 10-15 years show a moderate decline in real terms – but the reality is that the longer the projection span the more uncertain the projections become, which would call for additional prudence in the policy framework to manage natural resources revenue. Finite Resources The finite nature of resource revenues raises important issues of fiscal sustainability and intergenerational equity. As resource revenues come to represent a substantial portion of the national budget, a government may face pressure to increase expenditures.5 Recent discussions regarding 5 In a context of limited legal and institutional controls and powerful interest groups additional revenues flowing into the national budget may put considerable pressure on the government to increase recurrent spending, as various stakeholders each attempt to secure the largest possible share of the revenue windfall. This propensity is known as the voracity effect, and it frequent contributes to procyclicality and to unsustainable fiscal policies. See, Tornell and Lane, 1999. 16 government salaries in Mozambique6 illustrate the spending pressures and competition for public funds that will likely intensify as revenues from extractive industries rise. Increased spending should focus on investments in public goods and services that offer high returns—such as health, education and infrastructure—and such increases should always remain in line with the government’s projected long- term fiscal envelope. Spending decisions today may imply commitments in the future, such as the recurrent costs of investment projects or future obligations with public employees. Moreover, in determining how to manage resource revenues policymakers must balance the needs of current and future generations. Resources spent on the present population cannot be spent on generations of citizens to come, and the imperative to maintain equity in the returns to resource wealth is part of a broader principle of fiscal policy known as the inter-temporal budget constraint. The box below discusses how Botswana, a country known for its prudent management of natural resources, is nevertheless struggling to save an adequate share of its resource revenues. The fiscal policy framework therefore needs to take into account not only short-term macroeconomic stability but also the government’s long term fiscal stance. 6 In early 2013 healthcare workers presented a number of demands to the government, including a significant salary increase. These demands were initially not met and healthcare workers staged a strike that paralyzed Mozambique’s health services for several weeks. Ultimately the two parties reached an agreement to increase salaries by an average of 15 percent. This was reflected in the revised budget for 2013. 17 Box 1. Botswana: The Limits of Frontloaded Investment Botswana is often cited as a rare example of a developing country that has harnessed the power of its resource sector to generate lasting growth and sustainable improvements in the welfare of its people. Even though diamond mining is the largest contributor to Botswana’s GDP, the country ha s managed to avoid many of the problems frequently associated with resource-driven economies in Sub-Saharan Africa. The principles guiding the government’s spending and savings decisions include: (i) use mineral revenues to invest in human and physical capital, and ensure that adequate resources are available to meet new recurrent spending commitments in health and education; (ii) spend resource revenues through the budget and avoid off-budget expenditures or quasi-fiscal actions, (iii) maintain a countercyclical expenditure policy; (iv) respect expenditure-capacity constraints; (v) select public investment projects that have high rates of return; and (vi) save a portion of revenues to invest in financial assets. To help discipline expenditures and address the eventual depletion of its resources and their attendant revenues Botswana has been progressively implementing fiscal rules. The type of fiscal rule selected has changed over time to meet the demands of shifting policy objectives. Initially the “Principle of Sustainable Budgeting” was adopted to ensure resource revenues would be invested or saved instead of consumed. This led to the development of the Sustainable Budget Index, defined as the ratio of non-education, non-health recurrent spending to non-mining revenue, which had a fiscal target of 1:1. Later a fiscal rule was adapted to ensure spending was in line with medium-term government revenue, with a total spending target of 40 percent of GDP. Overall the expenditure-growth fiscal rule helped to discipline expenditures, but to date only a small portion of Botswana’s resource wealth has been saved and invested in long-term financial assets. Moreover, recurrent public spending has risen over time and is now consistently high, which is a concern as the country’s diamonds reserves are gradually being depleted. Revenues from diamonds are expected to stop growing sometime in the 2020s, with the expected exhaustion of reserves occurring around 2030 (Kojo, 2010). Figure 3: Total spending as a proportion of GDP is high in Botswana and is being financed by regular drawdowns of the Pula Fund Source: Bank of Botswana and Ministry of Finance and Planning Macroeconomic Impacts Beyond the public finances the growth of an export-oriented natural-resource sector can alter the national economy in complex and frequently adverse ways, often through a constellation of 18 macroeconomic impacts known as “Dutch disease”. Dutch disease refers to a situation in which an increase in resource exports generates large inflows of foreign currency, causing an appreciation of the real exchange rate. The rising exchange rate weakens the competitiveness of non-resource exports and domestic tradables, which results in slowing growth or even contraction in key non-resource tradable sectors such as agriculture and manufacturing. Meanwhile the resource industries, as well as a rapidly growing domestic nontradables sector, tend to attract the lion’s share of the country’s financial capital, further squeezing agriculture and manufacturing. A recent study (Ismail, 2010) argues that a 10 percent increase in oil revenues is associated with a 3.4 percent drop in value added in manufacturing. 7 Dutch disease can affect both developing and more advanced economies in similar ways, as illustrated in Figure 4 and Figure 5 below. The strong growth in the extractive sectors may have a negative impact on Mozambique’s macroeconomic stability. Strong growth in resource exports will increase the country’s vulnerability to terms-of-trade shocks. In Mozambique a decline in manufacturing—and especially in agriculture—could have a deeply negative impact on employment and income, given the large share of the population employed in these sectors and the highly capital-intensive nature of the extractive industries. An increasing concentration of economic activity in the extractable-resource and nontradables sectors could also inhibit Mozambique’s acquisition of new production capabilities through exporting, which is often an important source of knowledge spillovers and a driver of innovation. Figure 4: Australia experienced an appreciation of Figure 5: The appreciation of Angola’s REER is the REER, and a decline in manufacturing eroding the competitiveness of the economy Source: WDI and World Bank staff estimates Source: WDI and World Bank staff estimates Two effects lead to Dutch disease in resource-rich countries: the spending effect and the resource movement effect. The spending effect occurs as rising domestic income from the natural-resource 7 This finding is based on a sample of 15 countries: Canada, Columbia, Ecuador, Gabon, Indonesia, Iran, Kuwait, Malaysia, Mexico, Nigeria, Norway, Russia, Trinidad and Tobago, the United Kingdom and Venezuela 19 sector generates increased demand and spending by the public and private sectors, boosting nontradables (especially real estate, construction and services). To mitigate the spending effect the government can limit the growth of public expenditures. The resource-movement effect occurs as productive factors such as financial capital are shifted from the non-resource sectors to the extractive industries, exacerbating the decline of domestic tradables (Brahmbhatt et. al, 2010). These two effects present the government with a double bind that further underscores the importance of a balanced approach. Large-scale investment in infrastructure and human capital can boost the competitiveness of agriculture, manufacturing and other non-resource tradables, offsetting the impact of the resource-movement effect. However, higher rates of public investment will tend to compound the spending effect. Conversely, saving resource revenues in an SWF will slow the spending effect, but the resource-movement effect will continue unimpeded. The government must therefore carefully monitor developments in the non-resource tradable and nontradable sectors and take the shifting composition of the economy into account when making spending and savings decisions. Governance In addition to the macroeconomic and fiscal challenges described above, there is mounting evidence that strong public institutions and participatory decision-making systems are critical to effectively managing resource revenues. By providing an independent source of public income, one which is not tied to taxation, resource revenues can weaken accountability between the state and its citizens, encouraging authoritarianism, corruption, poor public service delivery and other negative governance consequences associated with the political culture of a ‘rentier state’ (Moore, 2004). There is a demonstrable correlation between the quality of public institutions and the growth rates of resource- rich countries (Mehlum et al., 2005 and Robinson et al., 2006). Figure 6 and Figure 7, below, illustrate this dynamic. The relationship between natural resources and growth is positive only for countries with strong governance; it is more ambiguous among countries with weak governance (World Bank, 2013). Figure 6: The share of natural resources in total Figure 7: The share of natural resources in total wealth and GDP growth among countries with wealth and GDP growth among countries with weak governance strong governance Source: World Bank staff estimates Source: World Bank staff estimates 20 B. Developing a Strong Fiscal-Policy Framework The previous section discussed some of the special characteristics of resource revenues that can pose challenges for a government attempting to manage them responsibly; this section is designed to help guide in the process of developing a fiscal-policy framework that is appropriate for Mozambique. The fiscal framework will help to determine the optimal pace and scale of public spending and saving in order to realize government, while also addressing the unique challenges of resource revenues. However, the impact of the fiscal-policy framework will depend not just on how well it is designed, but also on whether a robust and supportive institutional environment is in place. The international experience suggests that Mozambique would benefit from adopting a multifaceted policy approach to managing resource revenues. Figure 8 summarizes the types of decisions facing Mozambican policymakers and the factors that policy makers would need to consider in designing an appropriate fiscal policy framework. Determining how to spend resource revenues should be guided by the objectives of the overall fiscal-policy framework. Is the government concerned about limiting revenue volatility, building its capability to quickly increase public spending, managing the macroeconomic impacts of resource exports, or a combination of factors? The answers to these questions will inform the design of the policy framework8. Concerns about revenue volatility will require establishing mechanisms to smooth expenditures and avoid procyclicality. Concerns about expenditure capacity can be addressed by limiting expenditure growth. Mozambique should adopt a fiscal rule that helps authorities achieve the country’s objectives while addressing key concerns discussed. A successful fiscal-policy framework must account for several crucial factors. These include: (i) the size of the fiscal envelop and the respective shares of resource and non-resource revenues; (ii) the country’s development needs that can be met through increased public spending; (iii) the capacity of the implementing agencies to spend additional resources effectively; (iv) necessary measures to address revenue volatility, prevent unchecked expenditure growth and avoid procyclicality in fiscal policy; (v) the projected resource-revenue horizon and its implications for fiscal sustainability and intergenerational equity; (vi) whether public institutions are strong, reliable, free from undue influence and otherwise capable of ensuring the efficient, effective, transparent and accountable use of resource revenues, and (vii) expectations formed by politicians and the public on the use of natural resource revenues. 8 IMF (2013) argues in a similar vein that there are 4 main objectives to be considered in the fiscal framework for resource-rich countries: macroeconomic stability, managing volatility, sustainability/ exhaustibility and meeting development needs. Country circumstances regarding capital scarcity and the time horizon for revenues from natural resources will then determine the type of fiscal policy framework appropriate for each country. 21 Figure 8: A conceptual framework for fiscal policy in a resource-rich developing country Non-resource income Resource income Factors to consider: Total government income  Size of the resource envelop  Public expenditure needs Fiscal-policy framework  Public expenditure capacity  Fiscal volatility  Revenue-reserve Fiscal targets for achieving policy objectives horizon (e.g. ensure public spending is in line with expenditure capacity or  Institutional address volatility of resource revenues) strength  Expectations (Politicians, public) Fiscal rule or guideline to support the target (e.g. expenditure-growth rule or the structural budget-balance rule) Improving the Improving the government’s capacity to government’s capacity to manage public spending manage public savings Given its considerable development challenges, Mozambique’s fiscal policy framework should prioritize increased public investment… Mozambique should use a large share of the revenue from natural resources to frontload public investments. The returns on public investments in infrastructure or improved health and education services are likely higher than the returns that could be earned from financial assets. The objective of frontloading spending is to generate higher returns through accelerated growth than could be achieved through savings alone. Accelerated public spending could also benefit future generations, thereby promoting the sustainability of the income generated. …however, the extent to which this strategy will be successful will depend on the quality of increased public investment. Increased public resources, if not well spent, could fail to generate the expected resources. In this context PFM reforms are key to Mozambique’s long-term growth. While PFM systems are being upgraded increases in public spending must be limited to take into account the limited financial-management and expenditure capacity of public agencies. There are limits to how fast any public sector can grow while still continuing to use fiscal resources effectively9. In light of the already large size of the Mozambican public sector and the existing capacity in the public sector it is important 9 Moreno-Dodson (2013) discusses the relationship between fiscal policy and growth and suggests that there are diminishing returns to public investment 22 that policymakers keep these limitations in mind. Once the appropriate level of spending has been determined, additional resource revenues should be saved for the future. A large increase in public spending could also exceed the economy’s absorptive capacity, contributing to overheating and Dutch Disease. Large and fast increases in public expenditure can exceed the country’s absorptive capacity, leading to overheating and Dutch Disease. Growth in private investment and consumption are likely to drive Mozambique’s economy over the next decade, the fiscal policy stance should take this into account, possibly saving a share of natural resource revenues as local supply capacity is enhanced. Managing domestic absorption becomes more difficult because traditional fiscal indicators are not useful in assessing the country’s fiscal stance. Mozambique should therefore develop fiscal indicators that are more appropriate for resource-rich settings, such as the non-resource primary balance or the structural primary balance. Any indicators used to assess the country’s fiscal stance should be easily understood and in line with the country’s implementation capacity. Mozambique should adopt the Fiscal Sustainability Framework (FSF) which allows for scaling up public investments with the aim of accelerating growth in the non-resource sector.10 The FSF recognizes that natural resources are finite but instead of translating natural resources into financial wealth (to be saved in a sovereign wealth fund) it allows for increases in public investments, in line with the country’s absorptive capacity to transform natural resources wealth into other forms of wealth – the type of approach that has been followed by successful resource rich countries (World Bank 2013). Segura- Ubiergo et al (forthcoming) suggest that the FSF, allowing for a scaling up of public investment to meet development needs, is the most appropriate option for Mozambique, highlighting that implementation would be challenging and the need to urgently build the Government’s capacity in a number of areas. There are other fiscal policy frameworks that put emphasis on preserving resource revenues for future generations, but they seem less appropriate for Mozambique given the large development needs and potential for higher returns through increased public spending. The ‘bird-in-hand’ approach saves revenue from natural resources and invests in long-term securities, using the returns on investments to finance a steady stream of consumption. This seems most advantageous for countries with a limited stock of natural resources and a generally high standard of living. A modified permanent income hypotheses allows for some front loading of spending through a drawdown of natural resource revenues, which is compensated by reducing spending in future years. This is the approach that has been applied in Timor-Leste, but policy-makers argue that this modified approach is too constrained for Timor-Leste’s development needs, which would probably also be the case for Mozambique (see box 5 for further information on Timor-Leste’s experience). 10 This is elaborated in a recent publication: IMF (2012a) “Macroeconomic Policy Frameworks for Resource-Rich Developing Countries”, Washington DC. 23 Box 2: Coordination between fiscal and monetary policy As already discussed, macroeconomic management becomes more complex in resource-rich environments, with countries often affected by Dutch-Disease and increased volatility. In this sense, a close coordination of fiscal and monetary policy becomes crucial. Countries would need to assess the fiscal stance using non-traditional indicators (such as the non-resource primary balance) to assess to what extent fiscal policy is contributing to short-term macroeconomic stability or adding pressure on the country’s limited supply capacity . Once resource revenues start to flow, if the country spends all resource revenues (or a large share of it as proposed in the FSF) there are a number of possible monetary policy responses given Mozambique’s flexible exchange rate regime. Under a flexible exchange rate regime, the exchange rate would tend to appreciate. Interventions by the central bank to limit appreciation could lead to inflationary pressures (if interventions are not sterilized) or higher interest rates and crowding out of the private sector (if interventions are sterilized). Increasing exports and public spending may also lead to increased external appetite to invest in Mozambique – putting additional pressure on domestic assets and potentially creating a bubble. The discussion above highlights the benefits of close policy coordination as well as the need for a gradual approach to increasing public spending to be in line with the country’s absorptive capacity. Using Fiscal Targets to Achieve Policy Objectives Fiscal targets can support the goals of the fiscal policy framework while also addressing some of the key challenges of managing resource revenues. The objectives of Mozambique’s prospective fiscal targets may include being able to increase public spending while keeping it in line with macroeconomic stability and the execution capacity of public agencies, addressing the volatility of resource revenues. Expenditure-growth targets can be established to address capacity constraints. Setting targets for annual increases in public spending can help in managing popular expectations about the pace and scale of resource-revenue spending as well as encouraging a realistic view among public agencies regarding their prospective funding trajectories. As the capacity of these agencies to spend effectively is strengthened and investment levels are progressively increased these targets can be redefined; however, targets that can be changed too easily may lose their credibility. In addition, the expenditure- growth target could be combined with a target that addresses the volatility of resource revenues and short term macroeconomic stability, such as the non-resource primary balance11. Targeting a positive non-resource primary balance can help to decouple fiscal policy from the volatility of resource revenues and keep public spending in line with the country’s absorptive capacity. The fiscal framework should take additional issues into account through an assessment of long-term fiscal sustainability and structural resource revenues. Evaluating Mozambique’s long-term fiscal position requires a careful assessment of the reserve horizon as well as the nature of all obligations and liabilities incurred by the government. Distinguishing between the structural and cyclical components of commodity revenues is also important to encourage fiscal stability, with structural components being 11 The non-resource primary balance is equivalent to the primary balance minus (net) resource revenue, preferably measured as a share of non-resource GDP (Sharma and Strauss, 2013) 24 significantly less affected by volatility. This requires taking the non-resource primary balance and adding the structural component of resource revenues further insulating fiscal policy from resource-revenue volatility. See Box 3, below, for a discussion of how the structural-balance target has been applied in Chile. While deciding the type of targets that will guide fiscal policy it will be important to keep them simple to understand and in line with the government’s implementation capacity. The Role of Fiscal Rules Many countries have adopted formal fiscal rules or guidelines to help realize their fiscal targets. A fiscal rule can place a quantitative constraint on expenditures, promoting fiscal discipline in line with the objectives of the fiscal-policy framework. Fiscal rules that establish clear limits on spending can foster more realistic expectations and provide a clear, understandable policy goal (Wyplosz, 2005). An important advantage of fiscal rules is their ability to ‘signal’ the government’s commitment to fiscal discipline. Reliably adhering to these rules will strengthen the government’s fiscal credibility (Debrun and Kumar, 2007). Due to these and other significant benefits fiscal rules are becoming increasingly popular. As of end-2012 75 countries had one or more fiscal rules in place, 28 of which were advanced economies and 47 were developing countries (Bova et al., 2013). There are a number of good-practice principles to consider when designing a fiscal rule. The rule should be based on a simple and easily understandable set of fiscal variables with numerical targets and straightforward objectives. The government should have sufficient discretion to respond to shocks, and an ‘exceptional circumstances clause’ may allow for temporary deviations from the fiscal rule under certain specified conditions. The rule should be monitored and enforced through a well-designed legal and institutional framework, including an appropriate statutory basis, strong oversight mechanisms, clear-cut transparency and accountability provisions, and effective enforcement procedures. Finally, all fiscal rules should be grounded in a broad national consensus regarding their nature and purpose and a common understanding of their benefits (Kopits, 2001 and Dabán, 2011). The experience of fiscal rules in resource-rich countries has been mixed. Even a perfectly designed fiscal rule will prove ineffective if it lacks sufficient political support or if it is inadequately linked with the country’s PFM mechanisms (Corbacho and Ter-Minassian, 2013). For example, the government of Chad developed a rigid fiscal rule but failed to properly integrate it into the country’s PFM system. Due to capacity constraints and governance issues several budgets were put in place with separate cash- management processes and revenues earmarked for specific purposes. While the national budget remained subject to deficits and arrears, the fiscal rule generated savings in the oil fund at relatively low returns. With limited political support the oil-fund rule was abolished in 2006 (IMF, 2007). Key lessons from this experience include the importance of tailoring a fiscal rule to the country’s circumstances, allowing the right amount of flexibility, and ensuring a strong and enduring political commitment. The successful deployment of fiscal rules in countries such as Chile and Botswana involved a process of trial and error, during which the fiscal rule was adapted to suit the country’s objectives and institutional structure. In both countries fiscal rules were revised in response to important changes in 25 macroeconomic conditions, which helped to galvanize broad political support. In this context Mozambique could pilot a fiscal rule; the pilot phase would yield important lessons for improving the efficacy of the rule before its formal adoption, maximizing its chances for success. The fiscal rules should also be simple to understand, to allow a broader understanding of fiscal policy in the country, and in line with the country’s implementation capacity. Box 3. Chile: The Successful Implementation of a Fiscal Rule and a Sovereign Wealth Fund Chile is among the world’s leaders in rules-based fiscal policy. In the late 1980s the government adopted a price-based rule linked to its SWF. When the price of its chief export, copper, rose above a given threshold all excess revenue was deposited in the Copper Stabilization Fund. When the price of copper fell below a specified minimum withdrawals from the Fund were made to compensate for the shortfall (Blondal and Curristine, 2004). This policy was changed after the structural budget balance shifted to a deficit for the first time in a decade, highlighting the critical importance of separating the cyclical and structural components of resource revenues and of the budget as a whole. Even though addressing price volatility was still a major policy objective the government became increasingly concerned with improving its net-asset position in order to meet its considerable contingent liabilities and cover its fiscal deficit. To achieve these objectives the government set a structural budget-surplus target of 1 percent of GDP. Like the previous price-based rule the new structural-surplus rule was designed to insulate public spending from volatile copper prices while building long-term fiscal savings. Two SWFs were established to hold and manage excess revenues, the Pension Reserve Fund and the Economic and Social Stabilization Fund, the latter functioning as an automatic countercyclical fiscal stabilizer. The Central Bank of Chile manages these funds in consultation with an Independent Advisory Financial Committee (Fuentes, 2009). The structural-surplus rule is enforced with a limited measure of flexibility based on prevailing macroeconomic forecasts (IMF, 2012b). Mozambique does not necessarily need to adapt a fiscal rule similar to Chile’s, which may have different objectives and implementation capacity. What is interesting in this case is how the fiscal policy framework has adapted to the country’s changing circumstances. Figure 9: Under the structural-surplus rule expenditures are based on long-term revenue projections, and temporary windfalls do not influence current spending Figure 10: The countercyclical role of the sovereign wealth fund 25 Cumulative Withdrawals ESSF 20 US$ billion 15 10 5 0 26 Sep/08 Sep/09 Sep/11 Sep/07 Sep/10 Jun/08 Jun/11 Jun/07 Jun/09 Jun/10 Dec/09 Dec/11 Mar/07 Dec/07 Mar/08 Dec/08 Mar/09 Mar/10 Dec/10 Mar/11 Mar/12 Improving the Government’s Capacity to Manage Public Spending The proposed fiscal framework foresees increased public spending to meet Mozambique’s vast development needs. Further improvements in the government’s capacity to manage public funds will ensure that the increased spending generates the expected results. The extent to which increased spending succeeds in generating the expected returns will depend on how efficiently and effectively resources are allocated and utilized. The growth of extractive industries can generate expectations and pressures to increase spending. In any political system, and particularly in weak-governance environments, powerful groups and individuals may attempt to influence the allocation of public resources to benefit narrow private interests at the expense of broader national objectives. The limited progress in poverty reduction revealed by the latest household survey suggests the need to renew efforts to ensure that public spending reaches the poorest and most vulnerable. In order for Mozambique to realize the benefits of increased public spending efforts will be required to improve the allocative efficiency and quality of its spending. 12 Improvements in the government’s public-investment capacity and its ability to manage a significantly larger resource envelope are prerequisites for any further expansion in public spending. Figure 11, below, illustrates several options for improving the government’s expenditure capacity that will be discussed later in this section. Figure 11: Improving the government’s capacity to manage public spending Strengthening PFM systems is particularly important for resource-rich countries. Well-designed PFM systems are critical to: (i) ensure transparency in the extractive industries; (ii) promote sound budget processes and accountability mechanisms for spending resource revenues; and (iii) strengthen financial management procedures to address the special characteristics of resource revenues. Enhancing PFM is especially important for resource-rich developing countries, as they often have weaker PFM systems than their non-resource-rich counterparts and public spending is the main channel to translate natural wealth into other forms of wealth. 12 These and other issues surrounding the efficiency and effectiveness of public spending as well as options for allocating spending to different priorities will be explored in a World Bank Public Expenditure Review for Mozambique, which is expected to be completed in June 2014. 27 Despite impressive progress on PFM reforms, Mozambique will need to further strengthen PFM in areas that are key for managing natural resources revenues. Mozambique’s substantial progress in PFM reform has been demonstrated by its relatively high Public Expenditure and Financial Accountability (PEFA) scores (Umarji et al, 2011). While the design of the PFM reforms in Mozambique has been impressive, there is often a significant gap between design and implementation (Andrews, 2013). The implementation of PFM reforms should be sequenced according to the government’s broader PFM strategy. This should allow for more fundamental aspects to be strengthened and consolidated before embarking on other needed reforms such as medium-term planning and budgeting; Table 1, below, provides a brief description of some of the more fundamental PFM areas that need strengthening, with a brief description of the current performance in these areas. Table 1: Key Areas for Strengthening PFM in Mozambique BASIC DIMENSIONS OF PFM CURRENT PERFORMANCE RECOMMENDATIONS Budget presentation Inconsistent presentation of resource Disaggregate and present resource revenues revenues across budget documentation. consistently in all budget documentation. The non-resource primary balance is not The non-resource primary balance should be included in budget documents. included in budget documents. Medium-term planning No multi-year projection of resource revenues The medium-term fiscal framework should and budgeting in the medium-term fiscal framework. include resource revenues projections, with alternative scenarios and risk analyses. Debt sustainability analysis is done but no The medium-term fiscal framework should reporting on contingent liabilities. incorporate contingent liabilities. There is a multi-year Integrated Investment The Integrated Investment Plan should be Plan but this is not fully funded or integrated incorporated in the multi-year and annual with planning and budgeting processes. budget processes. Ongoing operational and maintenance costs of investments should be factored in. Budget execution, cash- There are frequent in-year adjustments and Undertake monthly budget and financial flow management and reallocations, contributing to a C+ in the PEFA programming. budgetary credibility assessment on Performance Indicator 16 ‘availability of funds for committed expenditure.’ Single Treasury Account procedures are being Efforts to consolidate the Single Treasury consolidated, as a number of accounts are Account should continue. being closed. Comprehensiveness and Annual accounts are produced and are publicly All fiscal operations should be reported and transparency available. However, ll fiscal operations are not published with public access allowed. published, such as extra-budgetary operations, contingent liabilities, earmarked revenues and own generated funds. External accountability External accountability received a C+ score in The scope of external audits should be the last PEFA, but this is on an upward trend expanded. Efforts should continue to enhance from previous assessments. the number and efficiency of judges to Audit reports are based on the annual increase the number of audits and avoid a accounts. These reports are publicly available backlog. and are submitted to the National Assembly. Further efforts are required to follow up audit The Supreme Audit Institution (TA) has recommendations and ensure reports are consistently fulfilled the requirement to scrutinized by the legislature. submit its opinion by the end of November each year. Source: World Bank Staff 28 Once the government has adequately reinforced its PFM fundamentals it should begin to shift toward a medium-term expenditure framework (MTEF), a policy tool which offers considerable advantages for resource-rich countries. An MTEF provides policymakers with a basis for managing risks and uncertainties from a medium-term perspective. This is particularly important for mitigating the volatility and procyclical spending bias associated with resource revenues. Moreover, a medium-term approach to planning and budgeting for capital investment projects, and in particular the inclusion of operational and maintenance costs in project budgets, will help to reinforce fiscal sustainability as public-investment levels increase. By encompassing fiscal risks, debt sustainability and contingent liabilities MTEFs promote a sound inter-temporal approach of fiscal policy (Ossowoski, 2013). Laying the groundwork for an MTEF should start with measures to enhance the medium-term fiscal framework; specific suggestions for strengthening MTFF are included in Annex 1. Specific measures to improve public-investment management (PIM) are critical to support the anticipated increase in infrastructure spending. Strong systems to manage public investments are positively related to national income levels (Figure 12), and recent analytical work (Gupta et al., 2011) shows that the quality of public investment is a key determinant of the relationship between investment and growth. The absence of a strong PIM system can result in inefficient project-selection processes, implementation delays, weak procurement practices, chronic cost overruns, indefinitely suspended or abandoned projects, and inadequate provisions for operational and maintenance costs. Mozambique’s PIM capacity is relatively weak. The IMF recently developed an index measuring the efficiency of PIM (Dabla-Norris et al., 2011), and Mozambique’s score is slightly lower than that of comparable countries (Figure 13). Project appraisal is an especially important element of PIM, as appraisal processes are critical for selecting projects with high economic and social returns, yet Mozambique’s score in this area is well below average. Reforms to improve the quality of project preparation and appraisal could greatly improve value-for-money in public investments, but ensuring effective PIM will require sustained efforts to improve every aspect of the system. 29 Figure 12: There is a positive correlation Figure 13: Mozambique’s PIM system ranks slightly between GDP and PIM capacity below the average for comparable countries Source: World Bank staff estimates Source: Dabla-Norris, et.al, 2011 The PIM system encompasses at least nine distinct steps, which function together to determine the overall quality, efficiency and impact of public investment. Effective PIM requires: (i) a clear strategy to guide investments policy, project development and preliminary project screening; (ii) a process for formal project appraisal; (iii) independent review of project appraisal; (iv) sound project-selection and budgeting procedures; (v) efficient project implementation; (vii) responsive adjustment mechanisms to address unforeseen challenges; (viii) reliable, sustainable operation of completed projects; and (ix) effective project evaluation methods, including processes for incorporating lessons learned into the design of subsequent projects (Rajaram et al., 2010). Efforts are currently underway to improve the process by which public investment projects are appraised, evaluated and selected in Mozambique. The ongoing and planned efforts will no doubt improve the Government’s capacity to manage public investments, but it will be years before a robust PIM system is in place and public-sector capacities have been adequately strengthened. In the previous sections the note suggests that to manage natural resource revenues Mozambique should adopt a policy framework that allows for a frontloading of investments – but any planned increase in public investments must take these capacity constraints into account and a more gradual approach may be warranted given the weaknesses discussed. Public investment management systems will need to be designed to reflect Mozambique’s preferences and capacities. There is no PIM system that fits all circumstances, but there are a number of pillars that are common in reform processes in countries that have advanced most in this area and which suggest a way forward for Mozambique. 30 Box 4: Pillars for a successful Public Investment Management System in Mozambique All successful PIM systems have the following pillars, and Mozambique has already started work in a number of them: Appropriate legal framework. The level of formalization varies across countries, but a key element in all legal frameworks is establishing that all public investments need to be evaluated before any financing is allocated or implementation started. Institutional and organizational strengthening . Projects need to be evaluated independently, guaranteeing professionalism in the process and avoiding conflicts of interest. The project evaluation should therefore be carried out by an entity that is independent from project proponents and the entity should have sufficient decision making authority to eliminate, suspend or delay investment projects that do not seem adequately prepared or in line with the country’s priorities. A continuous capacity building and training program to strengthen the technical capacity of teams preparing and evaluating projects. Training and capacity building should ensure the sustainability of the system, by training professionals in the appropriate methodologies and use of manuals. Preparation of a common and uniform set of norms, procedures, shadow prices and methodologies to guide the preparation and elaboration of public investment projects. A common set of norms contributes to the comparability of projects and therefore the ability to prioritize and select projects. It also contributes to the independence of project evaluators since there is a common set of norms and procedures to be used. Design and implementation of an integrated project database to prepare, evaluate, supervise and monitor projects. Such a database would enhance the system’s transparency and improve efficiency and communication between the different stakeholders. The public should also have access to the database (or some of the information in it) promoting accountability. Build into the system ex-post evaluation of projects during project implementation and operation. The ex-post evaluation should analyze whether assumptions made during project appraisal were adequate, assess the quality of services being provided and permanently inform the public investment management system through the analysis and assessment of projects implemented. Optimizing the impact of Resource Wealth through Increased Public Savings The fiscal policy framework suggested will in all likelihood result in savings from natural resource revenues that will need to be managed. In order to maximize the benefits of its natural-resource revenues Mozambique should consider combining moderate short-term increases in public spending with a medium-term strategy for national savings. Savings mechanisms can advance a number of critical objectives, by serving as a buffer against revenue volatility, keeping public spending in line with the capacity of implementing agencies, mitigating the macroeconomic impact of large expenditure increases 31 and promoting intergenerational equity in the returns to resource wealth. Figure 14, below, illustrates several options for managing savings. Figure 14: National savings mechanisms The specific policy objectives of the fiscal framework will also inform the strategy for national savings. The savings mechanism chosen and the rules governing its operation must reflect the country’s unique circumstances and the government’s policy goals. The following section discusses: (i) the objectives of different types of resource funds; (ii) the optimal strategic asset allocation for each type; and (iii) the institutional framework for domestic investments. Resource funds are an important instrument of fiscal policy in many resource-rich countries. There are two major types of resource fund, for stabilization and savings purposes. The primary objective of a stabilization fund is to address the volatility of resource revenues and its impact on both the budget and the national economy. When commodity prices are high excess revenues are deposited into the fund, and when prices are low withdrawals are made to cover the resulting budgetary shortfall. The resource- price thresholds and specific amounts deposited in and withdrawn from the fund are governed by a set of fiscal rules consistent with the objectives of the country’s broader fiscal-policy framework. A stabilization fund would therefore not have separate spending authorities but it is an integral part of the country’s fiscal framework. By contrast, the purpose of a savings fund is to preserve wealth for future generations, to accumulate financial assets to offset the depletion of natural-resource wealth, and/or to establish a stable source of income disconnected from commodity revenues. These functions are not necessarily inconsistent with those of a stabilization fund, and indeed the two systems can greatly complement one another. The objectives of the resource fund will determine how its resources are invested. Since stabilization- fund revenues must be available for withdrawal over the short-to-medium term the investment horizon is broadly reflective of an underlying commodity-price cycle, with investments typically focused on relatively low-risk, high-liquidity instruments. These often include high-grade fixed-income investments and sometimes public equity. Savings funds have much longer investment horizons and typically involve 32 riskier, less liquid assets. Investment decisions for savings funds are driven by portfolio-optimization strategies that emphasize diversification of risks and returns. Savings portfolios often include public and private equity, real estate and infrastructure, and fixed-income assets typically represent less than 40 percent of the overall portfolio. Figure 15, below, shows the cumulative returns for typical portfolios over long investment horizons. Although a more diversified savings portfolio generates significantly higher returns than a typical stabilization portfolio, over short periods diverse portfolios are subject to substantial market volatility—illustrating the key tradeoff between different investment strategies. Figure 15: Cumulative real returns of various portfolios (in US$, 1983-2013) Source: World Bank Treasury calculations based on data from Ibbotson Associates and Bloomberg Note: Indicative Stabilization Fund Portfolio: 90% US Government Bonds between 1-3yr/ 10% US Stocks; Indicative Savings Fund Portfolio: 60% US Stocks/20% US Treasuries/20% US Corporate Bonds; Globally Diversified Portfolio: 40% Developed Global Stocks/10% EM Stocks/5% EM Debt/5% Commodities/20% US Treasuries/20% US Corporate Bonds Whilst resource funds hold considerable promise, international experience has been mixed, and the advantages of resource funds appear to be largely contingent on the underlying quality of governance. Even a fund with a well-designed governance structure and professional investment management may fail to maximize the benefits of resource extraction. A recent study suggests that in countries with resource funds government spending is less sensitive to resource-revenue volatility, but the same study also found that countries which established resource funds already tended to be less vulnerable to commodity-price shocks (Davis et al., 2003). Research also indicates that resource funds with both stabilization and savings functions have been more successful in countries with a strong commitment to fiscal discipline and sound macroeconomic management (Fasano, 2000). 33 Resource funds entail a number of risks that must be weighed against their potential benefits, and governance is again a critical determining factor. There are a number of risks involved in resource funds, particularly in weak-governance environments. These include the potential for: (i) direct raiding of the fund, in which the fund’s rules are violated and its resources are used for purposes not originally intended; (ii) indirect raiding, in which unsustainable fiscal policies are encouraged and effectively underwritten by the fund; and (iii) inefficient fund management, in which suboptimal investments, in some cases motivated by political objectives or personal gain, reduce the value of returns. Moreover, the presence of these risks may lead to political disagreement about the usefulness of the fund as a policy tool or spur controversy over its governing rules and investment strategy, and political interference itself represents a significant threat to the effective operation of a resource fund. 13 International experience with resource funds highlights the importance of developing a strong institutional framework and a sound PFM system. This includes ensuring that the resource fund is well integrated with the budget process and that transfers between the resource fund and the budget are transparent and guided by clear fiscal rules, with limited discretion over the management of the fund, to avoid opportunities for misuse and waste. In addition, revenues from the fund should not be earmarked for specific purposes, but should rather fund the overall budget under a unified budget-execution process. A single resource fund is capable of serving different objectives by maintaining separate portfolios for stabilization and savings (IMF, 2012), and establishing a single fund with multiple functions is often preferable to multiple funds. Responsibilities for the management of the fund must be clearly defined, and its operations must be subject to external oversight. Finally, an integrated asset-liability management system should be established in which the balance sheet of the resource fund is consolidated with the government’s other financial operations.14 The design features of SWFs vary from country to country due to differences in political institutions, economic conditions and the technical capacity of public agencies. The central government is typically the legal owner of the fund. The parliament adopts the laws that establish its formal structure and the rules for its management. In most countries the Minister of Finance (or equivalent official) or an office within the executive branch (i.e. the Cabinet or the Council of Ministers) acts as the fund’s owner and sets mandates for the institution that manages the fund’s assets. The government must establish a clear and transparent division of roles and responsibilities between the funds governing bodies, particularly the owner of the fund and the manager of the investment portfolio. This is essential to minimize political interference and promote asset management based solely on technical expertise. In many countries the central bank manages the resource fund. In developing countries central banks are often regarded as the most appropriate institution to manage a resource fund, as they have the technical capacity to manage financial assets in concert with their monetary and exchange-rate 13 The effectiveness of funds in Venezuela and Oman for example has been substantially reduced by frequent changes in the rules governing their operations. 14 The Bank of Botswana is a good example of a central bank that provides transparent, regular and accessible information on a resource fund and other aspects of public finance, including asset-liability information: http://www.bankofbotswana.bw/index.php/content/2009110614010-annual-report 34 functions. This institutional setup has been used in resource-rich developing countries, including Botswana, Chile, Ghana, Nigeria, Russia, Timor-Leste, and Trinidad and Tobago, among others. The alternative is establishing an independent institution to manage the fund’s assets. However, this is only appropriate for countries with a high degree of administrative capacity—such as New Zealand, South Korea and Singapore, all of which use this system—and it should not be considered a viable option for Mozambique. Creating a dedicated fund-management institution entails significant challenges and risks fragmenting the country’s institutions and technical capacity; in developing countries this has resulted in reduced oversight and weakened controls, leading to a significant increase in corruption. To prevent undue political influence in the management of the resource fund it is necessary to limit the discretionary authority of policymakers. Simple, straightforward and enforceable rules are essential to avoid overspending and safeguard the integrity of resource-fund management (Humphreys and Sandbu, 2007). To this end the Santiago Principles provide a set of generally accepted good practices that can guide countries in the management of their resource funds. The Santiago Principles cover three policy areas. These include: (i) the fund’s legal framework, objectives and coordination with macroeconomic policies; (ii) its institutional framework and governance structure; and (iii) its investment and risk-management framework (see Annex 2 for details). These areas are closely linked; a clear legal framework and objectives underpin a robust institutional framework and governance structure, which in turn supports an investment strategy consistent with the resource fund’s policy objectives. See Box 5 for a discussion on how some of these principles were applied in Timor-Leste. Box 5. Timor-Leste: Combining Increased Public Spending with an SWF in a Context of Strong Transparency and Accountability In Timor-Leste revenues from oil and gas comprise 94% of government revenue (Revenue Watch Institute, 2011). A Petroleum Fund was established in 2005 with both stabilization and savings objectives and is managed by the Central Bank. All revenues from petroleum go into the Petroleum Fund. Withdrawals can be made to finance expenditures through the state budget, but only under certain conditions. The amount that can be withdrawn is linked to a fiscal rule called the “Estimated Sustainable Income” rule, a variation on the permanent income hypothesis framework. The country’s estimated sustainable income is a long -term projection that is updated annually and can change over time. In practice transfers from the Petroleum Fund sometimes exceed what is allowed by the fiscal rule, but these exceptions require parliamentary approval. Timor-Leste’s experience has generated a number of positive lessons for managing SWFs. First, the government has demonstrated an enduring commitment to transparency and accountability; formal reports and independent advice regarding the fund’s operations are available to the public. Second, a clear division of responsibilities among various actors prevent the abuse of authority. Thirdly, investment policies responded swiftly and effectively to international market shocks in 2008 (McKechnie, 2013). Despite these achievements, the fund’s long-term success will depend on the government’s sustained political will and commitment to transparency and responsible financial management. Furthermore, the authorities will need to continue their efforts to strengthen the capacity of public agencies to manage a larger resource envelope and reinforce public investment management systems to ensure that resource revenues channeled through the budget are utilized as effectively as possible. Many SWFs invest exclusively in foreign assets, but there is growing interest in investing in domestic assets, especially to finance long-term infrastructure projects. A number of resource-rich developing 35 countries have been examining alternatives for investing resource revenues domestically. Domestic investments could be used to promote economic diversification and offset Dutch-disease effects by boosting the competitiveness of key non-resource tradable sectors such as agriculture. In some cases the rate of return to investment in the domestic market may be higher than investment in foreign assets, particularly if positive economic and social externalities such as the impact on job creation are considered. Domestically held savings could be used to expand access to finance by small and medium enterprises or entrepreneurs that may otherwise face difficulty in obtaining credit from private financial institutions. A strategy to diversify risks would typically suggest investing abroad, since the value of domestic assets (and therefore the government’s ability to make withdrawals from the fund) would decline in the face of a slowdown in the domestic economy. A fund with a large share of its portfolio invested in domestic assets will probably not provide a reliable source of countercyclical financing. While a case may be made for investing a resource fund’s assets domestically, policymakers should carefully consider the most appropriate mechanism for doing so and the impact domestic investment will have on short-term macroeconomic stability. One significant source of risk for domestic investment is its potential to increase demand, particularly for domestic nontradables, in an economy that may already be overheating. This risk can be addressed by focusing on investments that improve competitiveness in the non-resource tradable sectors (especially agriculture and manufacturing) and by adopting a cautious, gradual approach to increasing investment that minimizes its contribution to domestic demand. But competitiveness-enhancing investments in public goods and services should be done through the budget and not a SWF. Domestic investments should also generate positive returns to avoid diluting the commercial mandate of the fund. A fund that invests domestically will therefore need to carefully balance the need to maximize returns while investing in sectors that are otherwise not well served by financial markets. Gelb et al. (2014) provide a detailed discussion on SWF and possibilities for domestic investment, highlighting the difficulties in balancing financial returns and economic benefits of domestic investments as well as the risks of using SWF to invest domestically. Establishing a National Development Bank Many countries are now using national development banks to investment domestically, but this strategy also entails its share of risks. Resource rich countries may use a development bank to channel a share of the resource revenues to the domestic economy. The main rationale for using resource revenues to capitalize a development bank is to address financial market failures. A development bank’s standard mandate is to provide credit and financial services, often on a long-term basis, to clients that may have trouble obtaining credit from the private financial sector. In practice this means that small and medium enterprises are often major beneficiaries of a development bank (de Luna-Martínez and Vicente, 2012). When assessing the merits of a development bank it is important to thoroughly analyze the rationale for this type of public intervention. Are there demonstrable market failures that a development bank could address? Would a development bank be preferable to other prospective options? What positive economic and welfare impacts would justify the use of scarce public resources in 36 this manner? What perverse incentives might it generate? How would the bank’s legal framework be organized? And what would be necessary to ensure the integrity of its operations? 15 In many countries development banks have failed, generating financial instability and often forcing the government to intervene at a considerable fiscal cost. Based on the international experience one of the main challenges faced by development banks is their limited capacity for risk management. Assessing the creditworthiness of prospective clients, determining the types of credit policies that should be followed and strengthening their capability to collect on loans or execute collateral all present serious challenges for development banks (de Luna-Martínez and Vicente, 2012). As a result many development banks end up with an unsustainable share of nonperforming loans and must be rescued at the government’s expense.16 A further challenge is the risk of political interference; effective controls must be established to prevent undue political influence in lending decisions (Allen and Vani, 2013). The Government of Mozambique has announced its intention to create a development bank to fill some of the gaps identified in the country’s financial sector. The bank could provide long-term financing for infrastructure development, facilitate access to credit for small and medium enterprises and promote the development of key economic sectors such as agriculture. However, it is unclear whether sufficient safeguards are in place to attenuate the risks associated with a development bank. Development banks are owned and controlled by the state, and in many cases they receive regular government transfers and a guarantee on debt and liabilities; unlike commercial banks they are not (and should not be) financed by deposits from the general public. Although they are encouraged to be financially self-sustaining, in practice this is difficult to achieve due to their policy mandates and government backing. Consequently, development banks have much in common with state-owned enterprises, and they tend to suffer from similar challenges. The international experience has yielded a number of best-practice principles that Mozambique should follow if it creates a development bank. The bank should have a clear policy mandate, including a well-identified target sector or market segment. It should have a sound corporate-governance structure, including an independent and professional board of directors, and managers must be able to operate free from political interference. Careful attention must be paid to risk-management procedures and the share of nonperforming loans on the bank’s balance sheet. Strong regulations and active supervision are essential to ensuring that the development bank advances its policy goals while posing minimal risks to either the stability of the financial sector or the government’s fiscal position. A Mozambican development bank could further the recent expansion of financial services by focusing on lingering gaps in the financial market, including lack of long-term financing and access to credit by small and medium enterprises. A development bank should strive to complement existing financial institutions and not compete with them. The international experience also shows that a strong 15 A policy note on the potential role of a development bank in Mozambique, as well as a discussion of its main mandate and implementation challenges, is currently being prepared by the World Bank (World Bank 2014). 16 Sherif, 2004, estimates that the fiscal costs of intervening in state banks in Eastern Europe after the collapse of the Soviet Union were over 25 percent for a number of countries in Eastern Europe, such as Bulgaria (38 percent), Macedonia (30 percent) or the Czech Republic (26 percent). 37 governance environment is crucial for the success of a development bank, and without that environment the risks for mismanagement are so substantial that the country may be better off attempting to address financial market failures through other means. Managing Assets and Liabilities The advent of resource revenues will significantly increase the complexity of public financial management. Revenues from extractive industries create their own set of challenges and opportunities for asset and liability management by affording access to new financial instruments. Due to its considerable stock of natural-resource wealth and projected future revenue streams Mozambique will have access to debt-financing options which would not otherwise be available, including credit collateralized against future revenues, loans to finance the government’s equity stake in extractive - industry projects or the use of public guarantees that will partially mortgage future income from extractive industries. To coordinate its increasingly sophisticated financial operations the Government of Mozambique should consider adopting a comprehensive sovereign asset and liability management policy that includes both public assets and the (implicit and explicit) liabilities of the government, including social programs, obligations arising from state-owned enterprises, guarantees provided to both public and private companies, pension funds, etc. Without a comprehensive accounting framework the accumulation of contingent liabilities and off-budget expenditures could jeopardize an otherwise balanced fiscal stance. The management of the government’s assets and liabilities should be governed by straightf orward and transparent long-term objectives. Formulating feasible objectives will require systematic assessment of fiscal sustainability, sensitivity analysis of the variables that influence the macroeconomic cycle and analysis of the structure and risk exposure of the government’s balance sheet. Setting clear goals for asset portfolios will allow asset managers to design efficient investment strategies with the proper investment horizons and risk tolerances. The planning and budgeting instruments currently used by the Government of Mozambique (such as the Budget Law, the Medium-Term Fiscal Framework or the recently adopted Medium-Term Debt Management Strategy) contain limited information on assets and liabilities, and this information is not consolidated in any single document. Presenting all asset and liability information in a clear and comprehensive manner would be a crucial first step towards enhancing transparency and accountability and addressing the challenges of a more complex public financial system. The Importance of Public Expectations and Support for Resource-Revenue Policies Countries that have made the most of their resource revenues have used proactive public engagement to manage expectations and build popular support. The anticipation of large resource- revenue inflows can generate a range of expectations and pressures. The latter frequently include efforts to increase public-sector salaries, boost public spending and increase support to special interests. Meanwhile, the potential misuse of public funds can encourage cynicism and popular discontent, as 38 citizens of countries with a history of weak governance may assume that the incoming revenues will be squandered through corruption, patronage and graft, even when no misuse of funds has occurred. This mix of hopeful and pessimistic expectations highlights the importance of public outreach, consultative processes and multi-stakeholder collaboration in formulating resource-revenue policies. These efforts should include both the current political leadership and the opposition, representatives of the private sector, civil society and special interest groups, and ordinary citizens. The risk of unmet expectations may be exacerbated if citizens and communities believe they are not receiving an equitable share of the benefits from natural resources. Popular frustration and a sense of unfairness or exclusion may contribute to political conflict. Given recent media interest in Mozambique’s natural resources, including extensive discussion of the revenues from capital-gains taxes, the arrival of multinational firms engaged in exploration, and early signs of economic overheating at the regional level, unrealistic expectations about Mozambique’s resource wealth and prospective revenues from the extractive-industries sector may already be developing. In Mozambique communities displaced in coal producing areas have voiced their concerns vocally and a recent escalation of the tensions between the Government and Renamo, with Renamo threatening to disrupt coal exports, are reminders that social unrest related to the development of the extractive sectors are a real possibility. There is substantial evidence that natural-resource exporters are more likely to experience violent conflict and civil unrest, and this risk is especially elevated in post-conflict countries. 17For example, it has been posited that one of the major contributors to the decades-long conflict in Aceh, Indonesia, was grievances over the distribution of resource revenues and jobs related to the exploration of gas in Aceh’s East Coast. The perceived grievances of the local population drove the success of the Free Aceh Movement in gaining local support and financing, which contributed to a prolonged violent conflict (Ross, 2005). Similarly, research shows that natural resources contributed to outbreak of civil war in Sudan, which has ultimately led to the formation of a secession state, South Sudan. Control over the country’s oil created tensions between different religious and tribal groups, leading to attacks on the oil pipeline and ultimately a full-scale war (Ross, 2004). Even when research identifies no clear causal link between the unequal distribution of resource benefits and violent conflict, rebel groups can gain traction by using ostensible grievances to justify their actions and gain legitimacy (Collier, 2006). Given recent political tensions in Mozambique this experience should underscore the importance of engaging with the public and managing expectations. Efforts to involve the public in discussions around natural resources revenue should focus on managing expectations and building support in a number of key areas. In Mozambique public outreach and consultation priorities should include: (i) communicating the message that significant increases in fiscal space are expected only in the next decade, not immediately, and only if LNG extraction goes ahead as planned; (ii) determining the objectives for the fiscal-policy framework and the target amounts to be spent and saved; (iii) deciding which fiscal instruments should be used to support these decisions; 17 This has been cited as a cause of secessionist movements in countries such as Angola, the Democratic Republic of Congo, Indonesia (West Papua and Aceh), Myanmar, Nigeria, Papua New Guinea, and Sudan, see Ross, 2004. 39 and (iv) identifying priorities for increased public spending and investment as part of a participatory budget process. Establishing a common understanding of the nature and limits of resource revenues, and building popular support for resource-revenue management, will enhance the credibility of Mozambique’s overall fiscal policy stance and may diffuse or mitigate potential political controversy. Box 6, below, discusses the experiences of two countries, Mongolia and Chile. These two countries built support across the political spectrum for the management of resource revenues, which was largely ‘depoliticized’ and allowed policy to be grounded on sound economic principles. Box 6. Mongolia and Chile: Public Engagement and Popular Support for Resource-Revenue Policies Early resource-revenue management policies in Mongolia were based on populist programs that increased public spending but involved limited public engagement. This led to fiscal and macroeconomic instability, as well as the raiding of resource funds. In response the authorities introduced a series of reforms to improve resource-revenue management with a strong focus on increasing public awareness regarding the challenges of managing revenues from commodities as well as the type of programs that would be financed by the revenues. In 2010 a Fiscal Stability Law was passed, and in 2011 the leaders of all political parties signed a memorandum to not compete through populist promises that would compromise the fiscal stability of the government or the objectives of the resource funds. In Chile several of the country’s salient political -economy features contributed to building support for responsible fiscal policy and resource-revenue management. These included: (i) widespread acceptance that fiscal responsibility would be the cornerstone of the country’s democratic transition; (ii) a political system that encouraged cooperation and negotiated solutions, (iii) limited influence by special-interest groups, (iv) support from the opposition for a fiscal rule that limited the growth of public spending, and (v) a powerful Executive with and a Budget Office staffed by high-caliber economists. In addition, the authorities invested considerable effort in promoting public understanding and ensuring transparency so that the structural balance rule would be widely understood by the public. Initially there was skepticism from the opposition and the public over the calculation of the fiscal rule, and as a result two independent expert panels were established by the government and the opposition, which increased public confidence. Information on fiscal policy and the management of the country’s resource funds continue to be widely accessible. Source: Bloöndal and Curristine, 2004, Fuentes, 2009, Ognon, 2013, Schmidt-Hebbel, 2008 The process of building public support begins with a strong and sustained commitment to transparency and accountability. Research shows that in many resource-rich developing countries one of the major obstacles to efficient allocation of resources is a lack of transparency. The Government of Mozambique has made considerable improvements in transparency, as reflected in its Open Budget Index score, which rose from 28 in 2010 to 47 in 2012. However, there are a number of areas in which further progress could be made. For example, budget documentation should include the declining value of natural-resource stocks and the contribution of resource revenues to the government’s net wealth (Dabán and Hélis, 2009). To facilitate better public understanding of these issues information on resource revenues should be included in the Citizen’s Budget. Strengthening external accountability (which scored relatively poorly in the most recent PEFA assessment) would require increased legislative oversight and greater emphasis on external audit processes. Ensuring transparency in the resource sector requires maintaining compliance with the evolving Extractives Industry Transparency Initiative standard. Mozambique is currently EITI compliant, but in 40 order to remain so the government must demonstrate continual improvements in line with the terms of the new standard adopted in 2013. Specific measures to further improve resource-sector transparency could include: (i) providing contextual information on the extractives sector, such as contribution to the economy (for example GDP and export levels), production levels, laws governing the extractives sector, and an overview of how revenues are recorded in government accounts and budget systems; (ii) expanding the scope of EITI reporting to include resettlement payments, and revenues and expenditures at the subnational level; (iii) disclosure of contracts and licenses that govern the exploration and exploitation of oil, gas and minerals – to include legal provisions, actual disclosure practices and any reforms that are planned or underway; (iv) information on the allocation and awards of new exploratory licenses and concessions; and (v) ensure the information is accessible and available to inform public debate in key fiscal policy decisions. These measures would also help to improve Mozambique’s score on the Resource Governance Index, under which the county is currently classified as ‘failing’ with a score of 37. Mozambique scores especially poorly on reporting practices, with an overall rank of 46 out of 58 countries (Revenue Watch Institute, 2013). II. DESIGNING A FISCAL-POLICY FRAMEWORK TO MANAGE NATURAL-RESOURCE REVENUES IN MOZAMBIQUE This section present a simulation of potential revenues from the coal and natural-gas sectors over the next 20 years, illustrating some of the challenges discussed in previous sections and exploring options for addressing them. This exercise assumes that Mozambique choses to frontload investments, spending a relatively large share of its incoming resource revenues on priority development objectives. The exercise includes: (i) anticipated revenue levels from the extraction of natural resources based on long-term price projections; (ii) a simulated fiscal rule that limits public expenditure growth; and (iii) estimated public savings levels and the expected returns on these assets. The simulation is based on a number of underlying assumptions regarding the available resource envelope and economic circumstances both domestic and external. The objective of the simulation is not to use these scenarios as planning tools, but rather to stimulate debate on resource-revenue management. Details on the data and the modeling of volatility are provided in Annex 3. The Estimated Fiscal Envelope from Both Resource and Non-Resource Revenues The resource-revenue estimates reflect the impact of uncertainty on any projected level of production. The government’s share of revenues from the production of coal and gas will be based on royalties, gas profits and corporate income taxes paid by resource firms. These estimates are based on the fiscal regime in place in early 2013. Historical prices and projections are used for coal, gas and financial assets to model uncertainty regarding future commodity prices and asset returns. Estimated production volumes are relatively conservative (Figure 16). For coal exports rail transport represents the main constraint. Production volumes assume the use of the current transport capacity of the Sena Line and the additional capacity that will become available from the Moatize-Nacala Line (through Malawi), bringing total coal-export capacity to 31 million tons annually. Natural gas production assumes that 4 LNG trains will be operational for the duration of the simulation period (until 2032); however, recent 41 press reports suggest that up to 10 LNG trains may be feasible, depending on global demand. It is therefore possible that exports of both coal and gas could be higher than assumed in this simulation. Until 2020 it is expected that natural resources revenue will be generated primarily from coal and onshore gas operations in the south of the country, but as LNG exports commence it is expected that government revenues will sharply increase in a short period, with LNG becoming the dominant source of revenue after 2020. Figure 16: Projected Production Volumes Sources: World Bank estimates based on ICG/GoM, 2013, and Rosenfeld (mimeo) Expected changes in global LNG markets are likely to increase volatility. Long-term LNG contracts have historically been linked to prevailing crude-oil prices and are typically characterized by lower levels of volatility than other commodities.18 However, expected developments in global LNG markets suggest that this is likely to change, which could result in higher price volatility. LNG prices have been indexed to crude oil based on the security of supply, but changes in the LNG market due to greater competition between sellers, more price-sensitive buyers, increasing energy deregulation, more gas-on-gas competition from new pipeline infrastructure, and increasing availability of spot-price based LNG exports may make it more difficult to justify price premiums (Ernst & Young, 2013). The price of coal and gas exports could therefore be subject to significant volatility (Figure 17). In addition to volatility, external risks add uncertainty to any long term projections of the volume of coal and gas as well as future energy prices. These external risks relate to both long-term energy demand as well as alternative energy supplies that could compete with Mozambique’s gas, some of the more obvious ones being uncertain demand from Asia and shale gas developments in the US and other countries. 18 Calculation of long-term LNG contracts are based on: (i) the oil-price reference benchmark, commonly calculated using Japanese-customs-cleared (JCC) crude prices, which represent the average monthly price of a basket of different crude-oil types imported into Japan; (ii) the price slope, which defines the relationship between oil and gas prices and is multiplied by the JCC; and (iii) the constant term, a fixed-price element with an implicit relationship to shipping costs, (Ernst & Young 2013.) 42 Key Underlying Assumptions for Estimates of Resource and Non-Resource Revenues  LNG exports will begin in 2020 with the construction of 4 LNG trains.  Long-term prices for coal and LNG will be anchored at US$140/ton and US$13.6/mmbtu, in nominal terms, respectively.  Non-commodity revenue growth will be 10.7% from 2012-2032, slower than the 2010-13 rate of 17%.  Royalties as a share of gross production will be 3% for coal and 2% for gas.  Profit gas will be the government’s share of the profit s from export sales. This is by far the largest component in expected government revenues.  Income Taxes are assumed to be 32% of projected profits for coal; income taxes on gas are assumed to be 16% of profits prior to 2025 and 32% thereafter Figure 17: Assumed Prices for Coal (US$/ton) and Natural Gas (US$/mmbtu) LNG starts Volatility Sources: World Bank estimates Based on estimated production volumes and prices there will be considerable volatility in future resource revenues. The total size of resource revenues will increase steadily, and by 2032 they could reach US$9 billion (90 percent from gas and 10 percent from coal), or 21 percent of total revenues. This assumes that non-resource revenues continue to grow at a somewhat slower pace than currently. The 95 percent probability range for resource revenues in 2032 is between US$4 billion and US$17 billion (Figure 18). Consequently the estimated fiscal envelop, including both resource and non-resource revenues, varies between US$36 billion and US$50 billion. This simulation indicates the magnitude of volatility that revenues could experience. Prudent fiscal management requires shielding government expenditures from resource-revenue volatility through mechanisms to smooth public spending and avoid procyclical fiscal policy19. 19 It should be noted that this simulation treats non-commodity revenues as deterministic, i.e. not subject to volatility. Both commodity and non-commodity revenues may be affected by the same shocks (e.g. a global crisis that affects growth and commodity prices), and by 2032 a larger share of the economy may be tied to the commodities sector through backward linkages, which means that a decline in commodity prices will have a more significant impact on the non-resource economy and therefore non-commodity revenues. 43 Figure 18: Revenues from coal and gas could amount to 21% of total revenues by 2032 60 Total Revenues (in USD Billion) 50 40 30 20 10 - 2013 2017 2021 2025 2029 Source: World Bank estimates The Impact of Fiscal Rules The pace at which public spending is increased should take into account capacity constraints of public agencies as well as the economy’s capacity to absorb large increases in public spending . Mozambique will have the option to put aside a share of the natural resource revenues for stabilization purposes and spend the rest through the budget. Given Mozambique’s significant development needs and historically constrained fiscal space, these additional resources could enable the government to enhance the provision of public goods and services. But there are limits to how much a government can provide, and how quickly it can ramp up public spending in a relatively short period of time, risking the waste of scarce resources if expenditure growth exceeds the absorptive capacity of the economy and the government. Public spending in Mozambique is already very high, reaching 33 percent of GDP in 2012. Lower- middle-income countries spend an average of around 18 percent, while countries in Sub-Saharan Africa spend an average of just under 24 percent. If Mozambique spends all of its incoming revenues, by 2032 public spending would consume 36 percent of GDP, a very rapid increase given the high rates of GDP growth expected over the period. The note suggests that a policy framework that allows for frontloading spending is the most appropriate for Mozambique, but this should also be in line with both the government’s capacity to manage a much larger resource envelope as well as the economy’s capacity to absorb an increase in public spending without compromising macroeconomic stability. Due to Mozambique’s large and fast-growing public sector expenditure-capacity constraints are a matter of serious concern. If Mozambique spent all of its incoming resource revenues the growth rate of public spending would increase to 9 percent annually in real terms for the next two decades. This compares with average growth in a group of peer countries over the last decade of 6 percent. An average growth rate of 9 percent instead of 6 percent would roughly double the size of the budget over the projection period, 44 putting considerable pressure on public-sector agencies. Measures of the quality of governance and the strength of public-investment systems in Mozambique raise serious concerns that such a significant expansion of public spending in the short-to-medium term would exceed the ability of implementing agencies to use resources effectively. Limiting expenditure growth through fiscal rules would allow Mozambique to address many of its most pressing development challenges, while safeguarding the quality of public spending and conserving a share of resource revenues for macroeconomic stabilization and/or future spending. Adopting a fiscal rule designed to limit increases in public spending would address both revenue volatility and expenditure-capacity constraints; this policy note illustrates the impact of two such rules. The first fiscal rule would tie the pace of expenditure growth to the nominal growth rate of the non-resource economy. Mozambique’s non-resource economy is projected to grow at around 14 percent annually in nominal terms. Limiting the growth rate of public spending to this level would result in a more gradual increase in total expenditures, which would be at around 33 percent of GDP by 2032 while generating an estimated US$40 billion in savings. The second rule would limit expenditure growth to the decade-average for a group of peer countries. Under this rule public spending would increase by 11.5 percent annually in nominal terms to 21 percent of GDP by 2032 and accumulate savings of US$120 billion. This would bring the Mozambican public sector much closer to the size of the public sector in other African countries, and to that of comparable low- and middle-income countries worldwide, while still allowing for a significant increase in spending. Figure 19 and Figure 20 below illustrate changes in public spending under these two rules compared to a counterfactual scenario in which no rule is adopted and all revenues are spent immediately. Figure 19: The growth of total public expenditures Figure 20: Public expenditures as a share of GDP under three different scenarios under three different scenarios Source: World Bank staff estimates Source: World Bank staff estimates The fiscal rules to be adopted should address also short-term macroeconomic stability objectives. The expenditure limits presented above are simplified versions of the type of fiscal rules adopted in many resource-rich countries. A fiscal rule should also assess to what extent the fiscal stance is in line with short-term macroeconomic stability. The non-resource primary balance (NRPB) for example assesses how public spending contributes to short-term macroeconomic stability or adds pressure on the 45 country’s limited supply capacity. In designing an appropriate fiscal rule for Mozambique these and other considerations will need to be taken into account (see IMF 2012 for a discussion on the design of fiscal rules). Projections for Public Savings Given the inherent uncertainty in commodity prices and export volumes the prospective amount of savings accumulated in a resource fund is highly variable. If the Government of Mozambique targets a nominal growth rate of 14 percent, it is assumed that commodity revenues in excess of the expenditure- growth target are deposited in a savings fund. Withdrawals from the fund are assumed to happen when commodity revenues for any given year are below the planned expenditure for that year. Assets are assumed to be invested in government bonds with a 2 percent real annualized rate of return. The fund’s balance by 2032 will be affected not only by the average level of commodity prices but also by their growth path, i.e., whether higher or lower prices happen in the beginning or at the end of the modeled horizon. Higher commodity prices at the beginning of the period will contribute to higher savings, which will increase the impact of investment earnings on the overall balance. The 90% probability range for the fund balance in 2032 is between US$2.5 billion and US$92 billion, with the expected (average) size at US$40 billion. Figure 21: The Projected Annual Net Savings Rate and Resource-Fund Balance Source: World Bank staff estimates The simulation exercise suggests that resource revenues will be relatively large but potentially volatile; if spent immediately they will likely strain the government’s expenditure capacity , the economy’s absorptive capacity and transmit significant volatility to the annual budget. The simulation does not attempt to provide a sufficiently accurate revenue projection on which to base actual expenditure planning. The uncertainties about prices, volumes and other macroeconomic variables are simply too high at this stage, partly due to ongoing changes in global markets that could affect the demand as well as the supply of energy in the long term, resulting in high margins of error that render 46 these estimates unfit for planning purposes. However, the exercise reveals important lessons that should inform policy discussions in Mozambique. Given its stated fiscal- and development-policy objectives the government should consider adopting a fiscal framework that (i) frontloads investments to address priority development objectives, (ii) uses a resource fund to mitigate revenue volatility and conserve a portion of resource revenues for macroeconomic stabilization, and (iii) employs a fiscal rule to limit expenditure growth to avoid overwhelming the expenditure capacity of public agencies and the economy’s absorptive capacity. III. CONCLUSION AND RECOMMENDATIONS Managing natural-resource revenue is inherently challenging, and many countries have failed to take advantage of the enormous opportunities offered by their resource wealth. Resource revenues are volatile by nature and can lead to procyclical fiscal policy and severe budgetary instability. Extractive industries often develop rapidly, and large increases in government revenue from these sectors can overwhelm the expenditure capacity of public agencies. Extractable resources are inherently finite, and unless the revenue from resource extraction is effectively invested in physical and human capital resource wealth will fail to generate permanent and sustainable increases in income. Moreover, using resource revenues primarily to finance consumption expenditures may prompt difficult fiscal adjustments in the future, and inadequate savings result in a lack of intergenerational equity in the returns to resource wealth. The quality of governance, the transparency of public financial management and the integrity of public institutions are key elements in the relationship between natural-resource revenues and long- run growth. Countries with strong governance structures have often been able to translate their natural-resource wealth into other forms of capital. But countries with a history of weak governance have suffered from the mismanagement of natural resources, leading to macroeconomic instability, declining competitiveness, ineffective poverty-reduction policies and unsustainable economic growth. The anticipated inflow of coal and natural-gas revenues toward the end of the current decade will present significant challenges for Mozambique. The simulation presented illustrates some of these challenges. Natural-resource revenues will represent a large and increasing share of total public-sector financing, and at an estimated US$9 billion in 2032 resource revenues will comprise 21 percent of total revenues. The actual value of resource revenues by 2032 is highly unpredictable, with a possible range of US$4-17 billion translating into a total annual resource envelope of between US$36-50 billion. These revenues will likely be subject to significant volatility, and unless measures are taken to address it this volatility will be transmitted to the budget. If all incoming resource revenues were spent, public expenditures would consume 36 percent of GDP by 2032, far higher than the 18 percent average for lower-middle-income countries or the 24 percent average for Sub-Saharan Africa. Mozambique’s public sector is already very large by the standards of comparable countries, and further increases in public spending could severely strain the government’s capacity to spend its resources effectively. These projections suggest that Mozambique will face serious challenges as it attempts to manage its natural 47 resource wealth, and meeting these challenges will require the development of new policy tools and governmental competencies. Mozambique should adopt a fiscal policy framework that allows a frontloading of public investments in line with capacity constraints. Given its pressing development needs a framework focused on frontloading investment while maintaining sustainability may be most appropriate for Mozambique. Factors to consider in developing this fiscal-policy framework should include the projected size of the resource envelop, the government’s priority spending requirements, the expenditure capacity of public agencies, the changing sectoral composition of the economy, the intrinsic volatility of resource revenues, the resource-revenue horizon, the strength of the institutional framework for revenue management as well as expectations being formed. Fiscal targets serve as a tool for achieving the objectives of the fiscal framework and for measuring progress on these objectives. Two fiscal targets that would be particularly relevant to Mozambique’s case include an expenditure-growth target, in which the government would establish a target or set of targets for annual increases in public expenditures, and a target for the non-resource primary balance, which would account for the impact of fiscal policy on macroeconomic stability. These targets are not mutually exclusive and can complement each other. A fiscal rule could help to keep the growth of an already large public sector in line with its administrative capacity, while a resource fund would insulate the budget against the volatility of resource revenues and allow a portion of present revenues to be reserved for future spending. Limiting nominal expenditure growth to 11.5 percent annually would result in the public sector accounting for 21 percent of GDP by 2032, comparable to the average size of public sectors in the region, as well as those of other lower-middle-income countries worldwide. A fiscal rule supporting this target could lead to the accumulation of US$120 billion in public savings. Restricting nominal expenditure growth to the 14 percent projected annual growth rate of the non-resource economy would result in a much larger public sector, at 33 percent of GDP by 2032, and generate total savings of US$40 billion. The fiscal targets and rules that are ultimately adopted must reflect Mozambique’s development objectives and evolving economic and institutional circumstances. While the government has enhanced its administrative capacity there are still considerable deficiencies in the delivery of public goods and services that must be addressed before increased public spending can deliver on its potential. Well-designed PFM systems are critical to: (i) ensure transparency in the extractive industries; (ii) promote sound budget processes and accountability mechanisms for spending resource revenues; and (iii) strengthen financial management procedures to address the special characteristics of resource revenues. Among the government’s top priorities should be improving public-investment management in order to meet its stated objective of increasing investment in infrastructure. Improving public-investment management should focus on strengthening institutional mechanisms for project screening and appraisal (including independent review), as well as project selection and budgeting processes, project implementation and adjustment capacity, operational efficiency and project-evaluation methods. 48 The fiscal framework and choice of instruments for managing resource revenues should reflect Mozambique’s policy objectives, capacity constraints and political economy. Ready-made systems or frameworks that have worked well in other countries will not necessarily be successful in Mozambique. However, the international experience does yield critical lessons for resource-management policy. The fiscal framework should be well integrated with the budget and the public financial management system, and simple, transparent rules should guide how resource revenues are spent through the budget. Revenues that exceed the government’s spending capacity could be saved in a resource fund, and the government could generate significant returns on these savings over time. A resource fund could promote macroeconomic stabilization and/or savings objectives, and the fund’s mandate should influence its investment strategy. In a stabilization fund investments typically focus on liquid assets and low-risk, low-return foreign securities. Savings funds, on the other hand, tend to have longer-term horizons and investments are often made in riskier, less-liquid foreign assets. Sovereign wealth funds usually invest in foreign assets, although resource-rich countries increasingly see a role for domestic investment. There will be significant pressure on Mozambique’s capacity to absorb increased public spending and foreign investment. The country’s limited capacity to absorb more investment and risk- diversification considerations will limit possibilities to use the sovereign wealth fund to invest domestically. Responsibilities for the management of the resource fund should be clearly defined, and rules governing deposits and withdrawals must be simple, straightforward and easily enforceable. As the financial operations of the Government of Mozambique continue to evolve, Mozambique will need to more carefully manage its assets and liabilities and include a consolidated balance in its budget documentation. Mozambique should consider adopting a comprehensive sovereign asset and liability management policy that includes both public assets and the government’s (implicit and explicit) liabilities. An accumulation of contingent liabilities and off-budget expenditures could jeopardize an otherwise balanced fiscal stance. Fiscal policy and the management of national savings should be closely coordinated with debt and liability management, and the management of public assets and debt should be governed by focused and transparent long-term objectives. The Government of Mozambique is currently considering founding a national development bank to expand access to financing in the domestic economy; however, this prospect should be treated with a great deal of caution given the risks involved. Many countries with relatively undeveloped capital markets have sought to channel resources to the domestic economy through a development bank. While a publicly capitalized financial institution with a mandate to address specific gaps in the financial market could contribute to more inclusive growth, the international experience underscores a number of important risks arising from non-performing loans and from these institutions’ typically limited ability to gauge creditworthiness and low capability to collect on loans or execute collateral. Furthermore there is a risk of political interference in the operations of a development bank, as the presence of such an institution creates opportunities for cronyism and graft. 49 Effective natural-resource policy requires managing popular expectations and building enduring consensus; this in turn requires a high degree of transparency and accountability in both the resource sector and public financial management. If expectations are unmet this could lead to grievances that could result in widespread dissatisfaction, social unrest or even outright conflict. Engagement with the public should focus on: (i) establishing realistic expectations regarding the quantity and timing of resource revenues, (ii) adopting a consultative approach to developing the objectives of the fiscal-policy framework, (iii) facilitating dialogue on the instruments used to support the fiscal-policy framework, and (iv) collaboratively identifying priorities for public spending in line with a participatory budget processes. Measures to enhance transparency and accountability could include increasing public access to key fiscal information on resource revenues, strengthening external-accountability mechanisms and achieving compliance with the new EITI standard. In conclusion, in order to effectively transform its natural resources into the basis for sustainable long- term growth the Government of Mozambique must determine how to maximize the returns to increased public spending and savings, as well as the institutional requirements to do this effectively. To this end the government should consider an approach that combines a controlled increase in public spending with the progressive accumulation of savings. A fiscal rule will anchor a stable, long-term strategy for spending and savings, but only if it is properly designed and reinforced by effective oversight mechanisms. Effectively utilizing sophisticated fiscal-policy tools requires informed decision-making, a strong and continuously improving institutional environment, and robust, participatory debate regarding the mechanisms and processes most suitable for Mozambique. Frontloading public spending and doing so effectively will require improving the overall quality of spending, by deepening PFM reforms and accelerating efforts to enhance the authorities’ public investment management capacity. 50 Annex 1: Suggestions for how the Medium Term Fiscal Framework (MTFF) can be improved Several improvements are required to the MTFF for this to be an effective basis for implementing an Medium Term Expenditure Framework. The MTFF should provide a medium-term forecast of the economy and public finances and a policy framework that sets a medium-term target for fiscal aggregates. The table below summarizes where improvements can be made to the MTFF based on the latest submission (covering 2014-2016). Table 2: Areas where the MTFF can be strengthened in line with good practice requirements Good practice requirements Areas where the MTFF can be strengthened (i) The macroeconomic outlook Review of underlying macroeconomic conditions and The underlying assumptions of how different sectors challenges contribute to GDP growth are not included. Macroeconomic outturn for the previous years, There is limited discussion on trends in revenue and analysis of recent trends expenditure performance. Updated macroeconomic forecast for the current year There is a medium-term resource envelope – but this and medium-term forecast for the coming three years does not have disaggregated revenue information. Implications for economic and fiscal policies and There is some discussion on implications of global strategies economic trends but this could include a discussion on trends in key commodities such as LNG and coal. Analysis of macroeconomic risks The scope of analysis is limited (e.g. no contingent liabilities). Budgetary implications are not discussed. (i) Fiscal policy and management Summary of fiscal developments, covering broad fiscal There is some assessment of macroeconomic trends, but aggregates for recent years there is limited discussion on fiscal aggregates. Assessment of the extent to which fiscal performance This is not included. in the budget year just completed was consistent with fiscal policy Analysis of key fiscal and macroeconomic problems There is some discussion of changes in the economy, but that fiscal policy needs to address it is not clear how the government’s fiscal policy will be formulated to address the issues identified. Fiscal targets for the short to medium term (explained This is not included. in relation to the budget) Discussion of whether there is a short-term departure No – partly because medium to long term fiscal from a medium-to-long term fiscal objectives objectives have not been clearly stated. (ii) Public spending priorities and provisional resource allocations Analysis of recent trends in public spending by Trends only provided by economic class and territory – economic class and function – as well as reasons for not by sector. Reasons for trends are not provided, trends and changes except for the impact of the floods. Key expenditure policies to be addressed over the The discussion of sectoral priorities is not linked to the medium-term strategic programs defined in the poverty reduction strategy – nor to expenditure policies. Impact that key government policies will have on There is some discussion on new policies, e.g. creation of resource allocations districts and municipalities, and salary increases – but the implications for fiscal sustainability are not discussed. Resource allocation implications covering ongoing Medium-term resource allocations have not changed in expenditure commitments and new allocations line with the expanding resource envelop. There is some information on strategic projects, but these are not linked to the Integrated Investment Plan, and the funding requirements are not stated. 51 Annex 2: Key Tenets of the Santiago Principles for Managing Sovereign Wealth Funds  Governance and accountability  A clear objective  A sound legal framework  Adequate reporting systems  Integration in domestic policy formulation  Appropriate coordination  Clear rules on funding and withdrawal  Incorporating SWF data into macroeconomic datasets  Management of the nation’s wealth  A clear investment policy  Diligence, prudence, and skill in investment practices  A robust risk management framework  Investment motivation  Public disclosure of policy purpose and governance framework, and relevant financial information  Refraining from pursuit of objectives other than maximization of risk-adjusted financial returns  Public disclosures of general approach to voting and board representation  Fair competition in markets  Respecting and complying with all applicable host country rules, laws, and regulations  Not seeking advances of privileged information  Impact on global imbalances and capital movements  Disclosure of relevant financial information  Description of the use of leverage or disclosure of other measures of financial risk exposure  Execution of ownership rights consistent with the SWF’s investment policy  A transparent and sound operational control and risk management system 52 Annex 3: Statistics for Scenarios 1 years 3 years 5 years 10 years 15 years 20 years Expected Revenues (in USD billion) 0.20 0.40 0.57 3.50 7.09 8.77 Expected fund size (in USD billion) 0.02 0.09 0.22 3.58 22.74 39.96 Median fund size (in USD billion) 0.00 0.01 0.02 2.82 20.00 34.77 Minimum fund size (90%) 0.00 0.00 0.00 0.15 7.00 8.13 Minimum fund size (95%) 0.00 0.00 0.00 0.00 4.53 2.47 Minimum fund size (99%) 0.00 0.00 0.00 0.00 0.91 0.00 Average net saving 0.02 0.03 0.04 0.35 1.43 1.77 Average total spending 0.18 0.79 1.66 8.46 20.06 45.74 Average spending 0.18 0.26 0.33 0.85 1.34 2.29 Average total contribution (in USD billion) 0.20 0.88 1.87 11.94 41.49 81.06 Average total withdrawal (in USD billion) -0.18 -0.79 -1.66 -8.46 -20.06 -45.74 Model description: Commodity prices and asset returns are modeled by using an autoregressive model. Its parameters are estimated from the historical data with imposed means, according to our long-term assumptions on commodity prices (140$/t for coal and 13.6$/mmbtu for gas) and asset returns (2% real annualized returns). 53 References Allen, R. and S. Vani (2013) “Managing Non-renewable Resource Revenues” in International Handbook of Public Financial Management Chapter 24, Allen, R, R. Hemming and B. Potter (eds.), Palgrave Macmillan Andrews, M. (2013) “The Limits of Institutional Reform in Development,” Changing Rules for Realistic Solutions: Getting Stuff Done Cambridge University Press Blöndal, J.R, and T. Curristine (2004) “Budgeting in Chile,” OECD Journal on Budgeting, 4(2) Brahmbhatt, M., O. Canuto and E. Vostroknutova (2010) “Dealing with Dutch Disease,” World Bank, Washington DC Bova, E., N. Carcena and M. Guerguil (2013) “Fiscal Rules and the Procyclality of Fiscal Policy in the Developing World,” IMF-World Bank Conference on Fiscal Policy, Equity and Long-Term Growth in Developing Countries, presented on April 21st-22nd 2013, Washington DC Collier, P. (2006) “Economic Causes of Civil Conflict and their Implications for Policy,” Oxford University. Corbacho, A. and T. Ter-Minassian (2013) “Public Financial Management Requirement for Effective Implementation of Fiscal Rules” in International Handbook of Public Financial Management Chapter 2, R. Allen, R. Hemming, and B. Potter (eds.), Palgrave Macmillan, forthcoming. Dabán, T. (2011) “Strengthening Chile's Rule-Based Fiscal Framework,” IMF Working Paper, IMF Washington DC Dabán, T. and J-L Hélis (2009) “A Public Financial Management Framework for Resource-Producing Countries,” IMF Working Paper, WP/10/254, Washington DC Dabla-Norris, E., J. Brumby, A. Kyobe, Z. Mills and C. Papageorgiou (2011) “Investing in Public Investment: An Index of Public Investment Efficiency,” IMF Working Paper, Washington DC Davis, O.J., J.A. Daniel, S. Barnet et al. (2003) “Stabilization and Savings Funds for Nonrenewable Resources” in Oil Producing Countries Chapter 11, Davis, J.M., R. Ossowski and A. Fedilino, (eds.), IMF, Washington DC Debrun, X. and M.S. Kumar (2007) “Fiscal Rules, Fiscal Councils and All That: Commitment Devices, Signaling Tools or Smokescreens?” IMF, Washington DC Ernst & Young (2013) “Global LNG: Will New Demand and New Supply Mean New Pricing?” Ernst & Young Global 54 Everhart, S., and R. Duval-Hernandez (2001) “Management of Oil Windfalls in Mexico: Historical Experience and Policy Options for the Future,” World Bank, Washington DC Fasano, U. (2000) “Review of the Experience with Oil Stabilization and Savings Funds in Selected Countries,” IMF Working Paper, Washington DC Fuentes, J.R. (2009) “Managing Natural Resources Revenue: The Case of Chile,” Oxford Centre for the Analysis of Resource-rich Economies, Oxford, UK Gelb, A., S. Tordo and H. Halland, (2014) “Sovereign Wealth Funds and Domestic Investment in Resource-Rich Countries: Love Me, or Love Me Not?”, World Bank Economic Premise No. 133 Gupta, S., A. Kangur, C. Papageorgiou, and A. Wane, (2011) “Efficiency-Adjusted Public Capital and Growth”, IMF Working Paper 11/217, IMF, Washington DC. ICF International (2012) “The Future of Natural Gas in Mozambique: Towards a Natural Gas Master Plan. Final Report,” commissioned by the World Bank on behalf of Government of Mozambique IMF (2007) “The Role of Fiscal Institutions in Managing the Oil Revenue Boom”, IMF Fiscal Affairs Department, Washington DC IMF (2009) “Mongolia: Request for Stand-By Arrangement – Staff Report; Staff Supplements; Press Release on the Executive Board Discussion; and Statement by the Executive Director for Mongolia” IMF, Washington DC IMF (2012) “Macroeconomic Policy Frameworks for Resource-Rich Developing Countries,” IMF, Washington DC IMF (2013) “Republic of Equatorial Guinea 2012 Article IV Consultation,” IMF, Washington DC Ismail, K. (2010), “The Structural Manifestation of the `Dutch Disease’: The Case of Oil Exporting Countries,” IMF Working Paper, Washington DC. Kojo, N. C. (2010) “Diamonds are not Forever: Botswana Medium-Term Fiscal Strategy,” Policy Research Working Paper, World Bank, Washington DC Kopits, G. (2001) “Fiscal Rules: Useful Policy Framework or Unnecessary Ornament?” IMF Working Paper, Washington DC de Luna-Martínez, J and C.L. Vicente (2012) “Global Survey of Development Banks,” World Bank, Washington DC McKechnie, A. (2013) “Managing Natural Resource Revenues: the Timor-Leste Petroleum Fund,” 55 London: Overseas Development Institute Research Report Mehlum, H., K. Moene and R. Torvik (2005) “Cursed by Resources or Institutions?” Norwegian University of Science and Technology Working Paper Series, Trondheim, Norway Moore, M, (2004) “Revenues, State Formation and the Quality of Governance in Developing Countries,” International Political Science Review 25 (3): 297-319 Moreno-Dodson, B. (ed.) (2013), “Is Fiscal Policy the Answer? A Developing Country Perspective”, Washington DC: The World Bank Ognon, K. (2013) “Sovereign Wealth Fund: The Case of Mongolia” presented at the High-Level Policy Seminar on Managing Revenues and Optimizing the Benefits of Coal and Gas Resources in Mozambique, hosted by the African Development Bank, February 27th-28th 2013, Maputo, Mozambique Ossowski R. (2013) “Managing Non-renewable Resource Revenues” in International Handbook of Public Financial Management Chapter 24, Allen, R., R. Hemming and B. Potter (eds.), Palgrave Macmillan Rajaram, A., T.M. Le and N. Biletska (2010) “A Diagnostic Framework for Assessing Public Investment Management,” World Bank, Washington DC Robinson, J.A, R. Torvik, and T. Verdier (2006) “Political Foundations of the Resource Curse,” Journal of Development Economics 79 Rosenfeld, D. (2012) “The Coal Mining Sector in Mozambique: A Simple Model for Prediction of Government Revenue”, unpublished mimeo Ross, M.L (2004) “What Do We Know of Natural Resources and Civil War?” Journal of Peace Research, Sage Publications Ross, M.L (2004) “How do Natural Resources Influence Civil War? Evidence from Thirteen Cases,” International Organization, Cambridge University Press Ross, M. L. (2005) “Resources and Rebellion in Aceh, Indonesia,” in Understanding Civil War: Evidence and Analysis, Volume 2: Europe, Central Asia and Other Regions, eds. Collier P. and N. Sambinis, World Bank, Washington DC. Schmidt-Hebbel, K. (2007) “Chile’s Growth and Development: Leadership, Policy-Making Process, Policies, and Results,” Central Bank of Chile, Working Papers, N° 507 Segura-Ubiergo, A, M. Poplawski-Ribeiro, C. Richmond, A. Veloz, G. Melina, V. Lledó and E. Palacio, (2013), “Fiscal Framework Considerations for the New Resource-Rich Environment,” IMF Fiscal Affairs Department, Washington, DC 56 Segura-Ubierto, A., M. Poplawski-Ribeiro and C. Richmond, forthcoming, “Fiscal Challenges of the Natural Resources Boom’ in International Monetary Fund, forthcoming, “Mozambique – Achievements and Challenges”, Occasional Paper Sharma, N. and T. Strauss (2013) “Special Fiscal Institutions for Resource-Rich Developing Economies: The State of the Debate and Implications for Policy and Practice.” London: Overseas Development Institute Research Report Sherif, K. (2004) “Fiscal Costs of State-Owned Banks in Selected Economies of Central and Eastern Europe”. In K. Liebscher, J. Christl, P. Mooslechner , & D. Ritzberger-Grünwald (eds.), The Economic Potential of a Larger Europe (pp. 203-216). Cheltenham: Edward Elgar Shi, Y., (2012) “Role of Infrastructure Capital in China’s Regional Economic Growth,” presented at the International Association of Agricultural Economists Triennial Conference, Foz do Iguaųcu, Brazil, 18-24 August, 2012 Tornell, A. and P.R. Lane (1999) “The Voracity Effect” American Economic Review, 89(1): 22-46 Umarji, M. C. Lees, J. Cavanagh, E. Palacio and A. Paulo, (2011) “Avaliação de Despesa Pública e Responsabilidade Financeira (PEFA) em Moçambique 2010,” MB Consulting, Maputo, Mozambique World Bank (2012) “Mozambique Social Protection Assessment: Review of Social Assistance Programs and Social Protection Expenditures,” World Bank, Washington DC World Bank (2013) “How Much Wealthier is Mozambique after the Recent Discoveries of Coal and Gas?” World Bank Mozambique Policy Note World Bank (2014) “Elements to consider when establishing the envisaged Development Bank of Mozambique” Wyplosz, C (2005) “Fiscal Policy: Institutions Versus Rules,” National Institute of Economic Review No. 191 57