81104 SEPTEMBER 2013 • Number 123 Resource-Backed Investment Finance in Least Developed Countries Håvard Halland and Otaviano Canuto The global financial crisis and shrinking aid flows have led to decreased availability of long-term debt finance for Least Developed Countries (LDCs), particularly for infrastructure. On the other hand, resource-related foreign direct invest- ment (FDI) in those countries has remained substantial. This note presents two models in which the natural resource wealth of LDCs has been used as a means to overcome the dearth of finance sources necessary for non-resource-related investments, and outlines country-specific factors that could tilt the balance between risks and opportunities to the latter. Mind the Investment Finance Gap over the last decade, been high enough to compensate private companies, many of them global multinationals, for the hur- The global financial crisis has dramatically strained the sourc- dles of operating in very adverse investment climates. es of traditional private long-term finance available to devel- Though the recent softening mineral commodity prices oping countries, particularly for infrastructure. In parallel, rules out more marginal mining projects, billion-dollar invest- aid flows have been diminishing. Although bond issuance has ments continue to pour into the sector, even under the most surged since the crisis, this has been geographically concen- difficult geographical and political circumstances, particular- trated outside the LDCs (Chelsky, Morel, and Kabir 2013).1 The Federal Reserve’s and other major central bank’s expect- ly in Africa. As a result, the LDCs have in fact been receiving ed eventual tightening of monetary policy is likely to further more FDI—as a share of gross domestic product (GDP)—than reduce the availability of long-term finance in countries with other more advanced developing countries (figure 1; Brahmb- the largest infrastructure deficit. hatt and Canuto 2013), and FDI to Africa has quintupled Many of the LDCs that lack access to international capi- since the turn of the millennium, from US$10 billion in tal markets, and where domestic capital markets are underde- 2000 to US$50 billion in 2012 (UNCTAD 2013). veloped, are also rich in natural resources (Canuto and Caval- So, one may ask, what kind of financing opportunities lari 2012). For investors, they represent an enticing but could result from the combination of weak governance en- challenging mix of potential high returns on investments in vironments, governments’ lack of capital market access, the extractives sector, with high political, macroeconomic, and natural resource abundance? Well, as it turns out, quite and project risk associated with poor policy climate and weak a few. Investors with a stomach for risk have invented fi- institutional capacity. To an important extent, governments’ nancing models that provide them with a competitive edge limited access to capital is a reflection of this perceived risk. In in this type of circumstance. Additionally, developing the extractives sector, nevertheless, expected returns have, country governments are seeking to become more proac- 1 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise Figure 1. FDI Inflows, 1980–2012 deals, also pioneered in Angola and later to become a main 6 all developing countries vehicle for financing that country’s postwar reconstruction. This financing model was later used in several other African Least Developed Countries 5 countries, predominantly by Chinese banks, including China Development Bank, but recently also by Korea Exim Bank for percent of GDP 4 the Musoshi mine project in the Democratic Republic of 3 Congo (DRC). According to Korea Exim Bank (2011), “the [Korean version of the RfI] model was strategically developed 2 to increase Korea’s competitiveness against countries which have already advanced into the promising market of Africa. 1 This agreement is the first application of the model.” Back-of- 0 the-envelope estimates based on publically available informa- tion indicate the value of signed RfI contracts in Africa to be 80 84 88 92 96 00 04 08 12 19 19 19 19 19 20 20 20 20 at least US$40 billion, most likely higher, although it is un- Source: Brahmbhatt and Canuto (2013). clear how many of these contracts have been fully implement- ed. Types of infrastructure projects have included railways, tive with regard to productive investment of rents generat- power plants, hospitals, and roads. Close to US$10 billion ed from extractives. worth of contracts were reportedly signed in 2011 and 2012 This note considers dominant trends in resource- alone, with at least US$6 billion in contract value under nego- backed financing for infrastructure. First, there is the con- tiation in 2013.4 tinuum from oil-backed lending, through resources for in- RfI deals, not to be confused with “packaged” resource- frastructure (RfI) deals, to possible options for government infrastructure deals—where the infrastructure is primarily bonds backed by future resource revenues. Next is another ancillary to the mine, such as rail mine-to-port links for ore recent but growing trend in several resource-rich developing transport—have been described as “swaps” or “barter” ar- countries, the use of their newly established sovereign rangements. However, RfI is better understood in terms of wealth funds (SWFs), not only for stabilization and inter- basic and familiar concepts from investment finance. Under generational saving through investment in foreign curren- an RfI arrangement, a loan for current infrastructure con- cy–denominated assets, but also for domestic investment, struction is securitized against the net present value of a fu- mainly in infrastructure. ture revenue stream from oil or mineral extraction, adjusted Resource-Backed Financing for risk. Loan disbursements for infrastructure construction usually start shortly after signing of the joint infrastruc- In Africa, the financing opportunities provided by high- ture–resource extraction contract, and are paid directly to risk, institutionally challenging, and resource-rich contexts the construction company to cover construction costs. The were first identified by Standard Chartered Bank in re- revenues for down payments on the loan, which are dis- sponse to the Angolan government’s demand for revenue to bursed directly from the oil or mining company to the fi- fund its war against Jonas Savimbi’s UNITA rebels. In the nancing institution, often come on stream a decade or more absence of a credible sovereign guarantee, given the Ango- later, after initial capital investments for the extractive proj- lan government’s low creditworthiness at the time, Stan- ect have been paid down. dard Chartered offered an arrangement whereby its lending The grace period for the infrastructure loan hence de- was to be guaranteed by future oil revenues. Other Western pends on how long it takes to build the mine or develop the banks soon followed suit, including BNP Paribas of France, offshore oil field (for onshore oil fields, the time line would be Commerzbank of Germany and others, and by the end of significantly longer if new pipelines have to be built), on the the war in 2002, the Angolan government had taken out 48 size of the initial investment, and on its rate of return. Large oil-backed loans (Brautigam 2011). According to Alves extractives projects can cost US$3–US$15 billion and take (2013), oil-backed lending remains a common format for 10 years or more from discovery to commercial operation. several banks that do business in Africa, a recent example Given the consequently long grace period for the infrastruc- being the Brazilian National Development Bank’s (BNDES) ture loan, getting the discount rate right, appropriately ad- loans to Angola.2 justed for risk, is essential. With the investor taking a signifi- Following the initial wave of resource-backed financing, cant share of operational, economic and political risk, this is a China Exim Bank in 20043 started offering so-called RfI nonrecourse loan, and an element of official or semiofficial 2 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise concessional finance to reduce investor risk has so far been a include strategic domestic investment. Examples include standard component of RfI deals. the newly established Nigeria Sovereign Investment Au- A “third generation” model of resource-backed finance, thority, and its Nigeria Infrastructure Fund, as well as the not yet developed, could consist of commodity-backed securi- Fundo Soberano de Angola. Others are in the process of be- ties (Songwe 2013). In this case, the resource-rich country ing created or are under discussion in Colombia, Morocco, would raise funds in international markets through the issu- Tanzania, Uganda, Mozambique, and Sierra Leone (Gelb et ance of a bond denominated in one of the reserve currencies. al. forthcoming). Payment obligations would be secured by future resource rev- Risks and Opportunities of enues, payable by the resource company into an offshore es- Resource-Backed Finance crow account, from which transfers to bondholders would be made, subject to appropriate investment guarantees. A common trait of the types of resource-backed investment finance described above is the very high risk if implemented Mobilizing Sovereign Wealth Funds for without due regard to the establishment of institutional and Long-Term Development Finance procedural safeguards. Before discussing these risks, however, Resource-rich developing countries were, until recently, en- a proper consideration is warranted of the characteristics that couraged to invest resource revenues according to some ver- have made such arrangements attractive to host country gov- sion of the permanent income hypothesis (PIH).5 Most of the ernments, and hence are generating demand. revenues would by this criterion be invested abroad in foreign A major benefit of resource-backed financing models to currency–denominated assets, so as to provide an even future democratically elected governments, or even nondemocratic revenue flow as subsoil reserves were depleted. Given the po- ones that need some sort of popular legitimacy, is that they tentially large economic and social externalities from domes- allow these governments to provide a return to citizens while tic investments in infrastructure, as well as potentially higher in office, and long before the extractive project is generating financial rates of return from domestic investment than that revenue or turning a profit. Additionally, in weak governance from long-term foreign assets, leading academics (Berg et al. contexts, RfI contracts can provide what Collier (2010) has 2012; Collier et al. 2009; van der Ploeg and Venables 2010) called a “new commitment technology,” whereby extracted have recently argued that, in countries with a large infrastruc- resources are with certainty offset by the accumulation of a ture deficit, it may instead be justified to front-load invest- productive capital asset. As Wells (2013) points out, this con- ments. The International Monetary Fund has also moved trasts with the frequent use of signing bonuses and royalties away from strict adherence to the PIH toward more invest- to fuel increased consumption, including higher public sector ment-focused policies. salaries. Hence “a wise Finance Minister may reasonably de- In a parallel development, SWFs based in high-income cide that this is much safer than letting the revenues flow countries have, over the last decade, increasingly been looking transparently into the budget and then hoping to emerge tri- to diversify into emerging markets. For example, funds from umphant from the subsequent political contest for spending” the Gulf countries have been estimated to hold 22 percent of (Collier 2010). RfI deals also reduce the risk of capital flight, their assets in Asia, North Africa, and the Middle East (San- by resource rents being transferred abroad (Lin and Wang tiso 2008). Over the same period and earlier, a number of forthcoming), as has frequently happened where resource SWFs in high- and middle-income countries implemented abundance and weak governance combine. Finally, in African policies that included domestic investment. Funds where do- contexts, the challenges of taxing and spending resource rev- mestic assets account for a significant share of total invest- enues efficiently may be “so daunting that governments find it ment include Singapore’s Temasek, New Zealand’s Superan- more advantageous to receive payments in kind” (Collier nuation Fund, Kazakhstan’s Samruk-Kazyna, and Malaysia’s 2013). Kazanah. Gelb et al. (forthcoming) list 14 SWFs that invest The risks of resource-backed financing are nevertheless domestically. substantial. Resource-backed loans and bond structures may, Recognizing the need for macroeconomic and fiscal in weak governance contexts, mortgage the nation’s subsoil buffers against highly volatile resource revenue flows, sever- wealth without much productive investment to show for it, al resource-rich developing countries have recently estab- thereby constraining future options for financing develop- lished, or are in the process of establishing, SWFs funded ment. There also may be implementation challenges. To at- from oil, gas, or mining revenues. Motivated by the trends tract investors, resource-backed bonds are likely to need a discussed above, respectively of investing in emerging mar- structure that isolates them from revenue variations arising kets and investing domestically, several of these new extrac- from highly volatile resource prices. Additionally, reflecting tives-based SWFs have defined or are considering mandates risks of sharp drops in resource prices, some sort of backstop that go beyond stabilization and intergenerational savings to will likely be necessary to provide the confidence needed by 3 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise investors that the government will not default on payment contribution to economic diversification. Finally, as has of- obligations if resource revenues fall. If the country has a low ten been the case with more standard types of contracts and sovereign credit rating, it may be difficult to establish a back- licenses in the extractives sector, RfI deals have frequently stop credible enough to ensure bond ratings are higher than been nontransparent, thereby increasing the risk of insuffi- the sovereign rating.6 cient public oversight and, for investors, the risk of future RfI deals, on the other hand, face challenges in terms of government demands to renegotiate the contract. This risk ensuring a proper valuation of the exchange, including ap- to investors is significant since, after the infrastructure has propriate discount rates and pricing of risk, and ensuring the been built, the government has an incentive to renege on quality of the built infrastructure during the construction the payment obligations associated with continued extrac- phase and after, as well as setting up proper operation and tion rights. maintenance structures in a low-capacity environment. To When it comes to domestic investment by SWFs, the address valuation and risk issues, assessment of the RfI op- main risks arise from the double role of the government as tion would need to start with estimates that compare infra- the owner of the fund as well as ultimately being responsi- structure costs with those that would arise from implemen- ble for its investments. This double role can undermine the tation by conventional fiscal and investment models, quality of investments as well as the wealth objectives of the whereby resource revenues would go into the budget and fund, in addition to potentially destabilizing macroeco- construction would be financed by the public spending sup- nomic management (Gelb et al. forthcoming). In low-capac- ported by these revenues. Collier (2010) has suggested that ity environments—with weak governance, public invest- proper valuation should take place through open competi- ment systems and regulatory frameworks, and where tion for the bundled contracts—although it is not yet clear coordination among public entities is lacking—such risk is how the frequently significant role of concessional finance magnified, and many resource-exporting countries have would be accounted for by an auction mechanism. The day implemented massive investment programs to little effect. of open competition for RfI contracts may be arriving, but so Publically owned wealth funds that invest domestically are far most proposals have originated from firms seeking op- nothing new, and public pension funds tend to have a sig- portunities either on the extractives or the infrastructure nificant proportion of their capital invested within the sides, and then partnering with other firms and a financing home country. However, when a fund receives its capital institution through an RfI to build a bankable deal to offer from resource revenues, as is the case with the “new” SWFs the government (Wells 2013). Ways of introducing competi- in developing countries, accountability is reduced since the tion could be based on processes established in some coun- fund essentially has “zero cost of capital” resulting from the tries to channel unsolicited infrastructure proposals into continuous stream of oil, gas, or mineral revenues. It does public competitive processes, thereby encouraging the pri- not need to raise capital in domestic or foreign capital mar- vate sector to propose potentially beneficial project concepts kets, and is not accountable to a group of interested stake- while maintaining the benefits of open tendering. Chile and holders such as pension contributors. In addition to this the Republic of Korea, for example, use a “bonus system,” agency risk, there is risk arising from investment mandates where a 5 to 10 percent bonus is credited to the original pro- that go beyond financial return to include social and eco- ponent’s bid in an open bidding round for the tender result- nomic externalities, and greatly complicate accountability ing from the unsolicited project proposal (Hodges and Del- because fund performance can no longer be benchmarked lacha 2007). on financial returns. Other relevant questions include: Is there an appropri- These risks can be managed, but not completely elimi- ate system in place to manage revenues generated after the nated. Domestic investments need to be screened primarily infrastructure investment has been paid down? Have prop- on the basis of financial returns, with allowance for home bias er measures have been taken to appraise, select, monitor, where there is market or close-to-market returns, and crowd- and evaluate the infrastructure projects, technically and fi- ing in rather than displacing private investors. Partnerships nancially? Since RfI substitutes infrastructure for fiscal with foreign SWFs or investment funds, with the home SWF flows, how will this affect debt sustainability? Have fiscal as a minority investor, would serve to reduce moral hazard and macroeconomic stability issues been properly consid- problems, in addition to opening up investment decisions to ered? As the loan component of RfI deals has been predomi- external evaluation and adding to the expertise at the fund’s nantly in the form of export credit, with labor and interme- disposal. Furthermore, meeting accepted international stan- diary goods imported from the funding country, potential dards for corporate governance, transparency, and audits is problems around macroeconomic absorption have been re- fundamental to ensuring fund independence and integrity duced. However, the extensive use of imports raises other (Gelb et al. forthcoming). In countries where a well-managed issues such as local employment, national value added, and development bank with a solid track record exists, invest- 4 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise ments with a home bias would benefit from being channeled can only bring full benefits if the government has the will through the development bank rather than fragmenting in- and capacity to establish the necessary procedural safeguards vestment decisions by setting up a separate structure. The for assessing, selecting, monitoring and maintaining the in- maximum domestic investment envelope, as well as home frastructure projects, as well as the competencies to credibly bias parameters such as a maximum allowable mark-down and forcefully negotiate the RfI contract based on a proper from the international benchmark rate of return, needs to be valuation and a more competitive tendering process, and if subject to parliamentary approval through the budget pro- all parties are willing to commit to transparency. Failure to cess. Where no home bias is defined, the challenge is to ensure build capacity on the government side, and implement the integrity of a process where domestic investments com- transparency, will likely result in low-quality, high-cost, bad- pete on equal terms with foreign assets, based purely on finan- ly selected, and poorly maintained infrastructure projects. cial return criterions. Since existing RfI deals have been underpinned mainly Notwithstanding the very substantial risks, potential ad- by export finance, procurement options for infrastructure vantages do exist. Taking advantage of its long-term horizon, construction have been limited to firms from the funding the domestic SWF is in a position to offer a range of instru- country. Oil-backed loans or bond issuance, on the other ments to share risk and make potentially attractive projects hand, provide the government with discretion in contracting commercially bankable. In some circumstances, and assum- with oil, mining, and construction companies freely, allowing ing that any home bias in the SWFs mandate can be clearly for competitive tendering of infrastructure as well as extrac- defined, the SWF could accept a somewhat below-market re- tive projects, but may represent an incentive for increased turn on domestic investments with large economic externali- consumption rather than investment if a strong commitment ties (Gelb et al. forthcoming). For example, instead of an ex- to investment does not exist. Furthermore, bonds may be a ternal rate of return of, say, 4 percent in real terms over an feasible option only if the country has sufficient creditworthi- investment horizon of 10 years, it could stipulate a real return ness to provide a credible backstop in the case of a shortfall in of 2 percent over a horizon of 20 years. resource revenues, or appropriate credit guarantees can other- wise be provided. Conclusion There could also be combinations of the different financ- Although prices of most minerals and fuels have fallen since ing options. For example, a SWF portfolio allocation invested peaking in 2008 (crude) and 2011 (metals and minerals abroad in foreign currency–denominated liquid assets could [World Bank 2013]), they are likely to remain far above his- be used to generate the backstop necessary to credibly issue torical averages due to increased demand from emerging mar- resource-backed bonds. Returning to the original point of this kets, primarily China and India. Resource-backed finance note, the success and failure of much debated resource-backed models are, in this context, likely to maintain or increase their financing models will be determined not on the basis of cate- share of infrastructure finance. gorical assessments of each model as “good” or “bad” for devel- Several financing options are available and will be de- opment outcomes, but on their deployment as it plays out in ployed according to the particular country and governance different country and institutional contexts, and on the insti- context. Allowing domestic investment by SWFs carries very tutional and procedural safeguards established to ensure significant risks, and precedents indicate that unless very proper governance. This is consistent with the current con- strong safeguards exist, failure is the likely result. There are sensus among economists, that whether natural resource huge moral hazard issues that may be impossible to efficiently abundance proves to be a “curse” or a “blessing” mostly de- address in a weak governance environment. In the best case, pends on circumstances and policies (Brahmbhatt, Canuto however, such investment may crowd in private sector inves- and Vostroknutova 2010; Ledermann and Maloney 2007; tors to borderline investments that would not be indepen- Canuto and Cavallari 2012). In the end, it all depends! dently bankable without some sharing of risk. By letting po- Acknowledgment tentially attractive domestic investments compete with foreign assets for investable funds, based on expected returns, The authors are grateful to Bryan Land and Silvana Tordo for competition for resource allocations based on market princi- their insightful comments on earlier drafts. ples, managed by capable investment managers, replaces com- About the Authors petition for resources through the budget process. If there is a perception that a SWF is likely to be raided Håvard Halland is a Natural Resource Economist for the Poverty by the next government in power, the current government Reduction and Economic Management (PREM) Network. Otavi- may prefer the earmarking of funds to investment implicit ano Canuto is the Senior Advisor on BRICS in the Development in RfI, rather than saving in a SWF beyond what is necessary Economics Department of the World Bank. He previously served for stabilization (Gelb forthcoming). On the other hand, RfI as the Bank’s Vice President and Head of the PREM Network. 5 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise Notes ———. 2013. “Can Africa Harness Its Natural Resources? Op- portunities for the DRC.” Paper presented at the World Bank 1. “The Least Developed Countries (LDCs) are 48 countries Kinshasa Conference, June. flagged by the United Nations Economic and Social Council Collier, Paul, Frederick van der Ploeg, Michael Spence, and Anthony as meeting certain thresholds for small population, low per J. Venables. 2009. “Managing Resource Revenues in Developing Economies.” OxCarre Research Paper 15, Oxford, UK. capita income, weak human development and high economic Congo, Democratic Republic of (DRC). 2007. “Joint Venture vulnerability to shocks… About two thirds are in Sub-Saharan Agreement, Group Gecamines-Consortium of Chinese Africa with the rest mostly in the Asia and Pacific region” Enterprises.” Ministry of Mines, http://mines-rdc.cd/fr/index. (Brahmbhatt and Canuto 2013). php?option=com_content&view=article&id=165&Itemid=126. 2. Oil had already been used as collateral in a previous agree- ———. 2008. “Cooperation Agreement, Democratic Republic of ment between Angola and Brazil for partial debt relief. Congo-Company Corporation Sinohydro.” January. Ministry of Mines, http://mines-rdc.cd/fr/index.php?option=com_content 3. Although the RfI model had been used by China Exim &view=article&id=165&Itemid=126. Bank on two occasions previously, in the Republic of the Con- ———. 2012. “Memorandum of Agreement, Mineral Resources Min- go (2001, US$280 million) and Sudan (2001, US$128 mil- ing SPRI-National Society of Railways of Congo SARL.” Minis- lion [Foster et al. 2008]), RfI is generally considered to have try of Mines, http://mines-rdc.cd/fr/index.php?option=com_co been pioneered through the far larger contract in Angola, and ntent&view=article&id=165&Itemid=126. Foster, Vivien, William Butterfield, Chuan Chen, and Nataliya has also been dubbed the “Angola mode” of contracting. Pushak. 2008. Building Bridges: China’s Growing Role as 4. For an overview of RfI projects in Africa, see Alves (2013) Infrastructure Financier for Africa. Trends and Policy Options and Foster et al. (2008). (PPIAF), Washington, DC: World Bank. http://siteresources. 5. “A commonly used benchmark for fiscal policy in a natural worldbank.org/INTAFRICA/Resources/Building_Bridges_Mas- resource–rich economy is the permanent income rule. Under ter_Version_wo-Embg_with_cover.pdf. this rule the country should save all resource revenues over Gelb, Alan. Forthcoming. “Comments on Halland, Håvard, John Beardsworth and James A. Schmidt (forthcoming), “Resource and above a certain permanent¬ly sustainable increase in the Financed Infrastructure: Origins and Issues.” World Bank, level of consumption, which is equal to the annuity value of Washington, DC. the country’s natural resource wealth.” (Brahmbhatt, Canu- Gelb, Alan, Silvana Tordo and Håvard Halland, with Noora Arfaa to, and Vostroknutova 2010, 115). and Gregory Smith. Forthcoming. “Mobilizing Sovereign 6. Xavier Cledan Mandri-Perrott, comments on Songwe Wealth Funds for Long-Term Development Finance: Opportu- nities and Risks.” Paper written for the G20 Working Group on (2013). Investment Finance. References Halland, Håvard, John Beardsworth, and James A. Schmidt. Forth- coming. “Resource Financed Infrastructure: Origins and Issues.” Alves, Ana Christina. 2013. “China’s ‘Win-Win’ Cooperation: World Bank, Washington, DC. Unpacking the Impact of Infrastructure-for-Resources Deals in Hodges, John T., and Georgina Dellacha. 2007. “Unsolicited Infra- Africa.” South African Journal of Foreign Affairs 20 (2): 207–26. structure Proposals: How Some Countries Introduce Competi- Berg, Andrew, Rafael Portillo, Shu-Chun S. Yang, and Luis-Felipe tion and Transparency.” PPIAF Working Paper No. 1. Zanna. 2012. “Public Investment in Resource-Abundant Devel- Korea Eximbank. 2011. “Resource Development in DR Congo oping Countries.” IMF Working Paper 12/274. through Water Supply Pipeline Construction.” http://www. Brahmbhatt, Milan, and Otaviano Canuto. 2013. “FDI in Least koreaexim.go.kr/en/bbs/noti/view.jsp?no=9671&bbs_code_ Developed Countries: Problems of Excess?” Global Finance id=1316753474007&bbs_code_tp=BBS_2. Mauritius n.1: 79–82. Ledermann, Daniel, and William F. Maloney, eds. 2007. Natural Brahmbhatt, Milan, Otaviano Canuto, and Ekaterina Vostroknuto- Resources, Neither Curse nor Destiny. Wasington, DC: World va. 2010. “Natural Resources and Development Strategy after Bank Publications. the Crisis.” In The Day after Tomorrow: A Handbook on the Future Lin, Justin Yifu, and Yan Wang. Forthcoming. “Contractual Innova- of Economic Policy in the Developing World, ed. Otaviano Canuto tion Is Needed for Structural Transformation.” Comments on and Marcelo Giugale. Washington, DC: World Bank. Halland, Håvard, John Beardsworth and James A. Schmidt Brautigam, Deborah. 2011. “The Dragon’s Gift: The Real Story of (forthcoming), “Resource Financed Infrastructure: Origins and China in Africa.” Oxford: Oxford University Press. Issues.” World Bank, Washington, DC. Canuto, Otaviano, and Matheus Cavallari. 2012. “Natural Capital Santiso, Javier. 2008. “Sovereign Development Funds: Key Financial and the Natural Resource Curse.” Economics Premise 83, World Actors of the Shifting Wealth of Nations.” Revue d’Économie Bank. Financière 9 (1): 291–315. Chelsky, Jeff, Claire Morel, and Mabruk Kabir. 2013. “Investment Songwe, Vera. 2013. “From Bottom Billion to Top Trillion: Using Financing in the Wake of the Crisis: The Role of Multilateral Commodity-Backed Securities to Support the Future of Africa’s Development Banks.” Economic Premise 121, World Bank. Resource Economies.” Unpublished memo, World Bank. Collier, Paul. 2010. “Principles of Resource Taxation for Low-In- UNCTAD (United Nations Conference on Trade and Develop- come Countries.” In The Taxation of Petroleum and Minerals, ed. ment). 2013. “Time Series on Inward and Outward For- Philip Daniel, Michael Keen, and Charles McPherson. London eign Direct Investment Flows, Annual, 1970–2012, Data and New York: Routledge. Compiled by the Financial Times August 19, 2013, “Offshore 6 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise Centres Race to Seal Africa Investment Tax Deals.” http:// Financed Infrastructure: Origins and Issues.’” World Bank, www.ft.com/intl/cms/s/0/64368e44-08c8-11e3-ad07- Washington, DC. 00144feabdc0.html. ———. 2013. “Infrastructure for Ore: Benefits and Costs of a Not- Van der Ploeg, Frederick, and Anthony J. Venables. 2010. “Absorb- So-Original Idea.” Columbia FDI Perspectives, Perspectives on ing a Windfall of Foreign Exchange: Dutch Disease Dynamics.” Topical Foreign Direct Investment Issues, No. 96, June 3, Vale OxCarre Research Paper 52, Oxford, UK. Columbia Center on Sustainable International Investment. Wells, Louis T. Forthcoming. “Comments on Halland, Håvard, World Bank 2013. “Commodity Markets Outlook.” Global Economic John Beardsworth and James A. Schmidt, ‘Resource Prospects, No 2, Washington, DC, World Bank. The Economic Premise note series is intended to summarize good practices and key policy findings on topics related to economic policy. They are produced by the Poverty Reduction and Economic Management (PREM) Network Vice-Presidency of the World Bank. The views expressed here are those of the authors and do not necessarily reflect those of the World Bank. The notes are available at: www.worldbank.org/economicpremise. 7 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise