84016 JANUARY 2014 • Number 133 Sovereign Wealth Funds and Domestic Investment in Resource-Rich Countries: Love Me, or Love Me Not? Alan Gelb, Silvana Tordo, and Håvard Halland “Innovation and best practices can be sown throughout an organization—but only when they fall on fertile ground.” Marcus Buckingham Sovereign wealth funds (SWFs) represent a large and growing pool of savings. An increasing number of these funds are owned by natural resource–exporting countries and have a variety of objectives, including intergenerational equity and macroeconomic stabilization. Traditionally, these funds have invested in external assets, especially securities traded in major markets. But the persistent infrastructure financing gap in developing countries has motivated some governments to encourage their SWFs to invest domestically. Is it appropriate to use SWFs to finance long-term development needs? Does it matter whether such investments are domestic or foreign-held assets? This note considers these issues, particularly the controversial question of using SWFs to finance domestic projects, motivated partly by SWFs’ perceived importance for development. Largely Uncharted Territory existing infrastructure and low-risk, new, bankable infra- structure projects in Europe and Asia. The motivation for The relevance of SWFs in investment financing and market these investments has been mostly commercial (Balding stability was underscored by the recent global financial cri- 2008). sis. Usually funded through excess foreign currency reserves, Yet, SWFs investing domestically are not as unusual as these funds have a variety of objectives and mandates, rang- one might expect. According to Truman (2011), domestic ing from addressing the macroeconomic impact of revenue holdings constituted 16 percent of total investments in a sam- volatility in resource-rich countries to ensuring intergenera- ple of 60 SWFs, although these included some pension funds, tional equity, addressing future financial needs, and protect- and Gelb, Tordo, and Halland (forthcoming) list 14 SWFs ing a country’s economy from extraordinary shifts in its fis- that invest domestically. But domestic infrastructure invest- cal situation. ment remains uncharted territory for most SWFs. In light of SWF holdings traditionally focus on external assets, prin- the pressing infrastructure needs, several resource-rich devel- cipally securities traded in major markets to respond to steril- oping countries have established, or are in the process of es- ization, stabilization, and risk/return objectives. Investment tablishing, SWFs with an expanded role as a national investor. in infrastructure projects is not uncommon in SWFs portfo- Angola, Mongolia, Nigeria, and Papua New Guinea are lios with long-term investment horizons. But most SWF in- among the most recent examples of this apparent trend. Ex- frastructure portfolios focus on nondomestic, high-return perts suggest that 20 SWFs are already mandated to invest 1 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise domestically (Monk 2013), and more are in the making, for tic infrastructure investment in developing economies is not example, Colombia, Morocco, Mozambique, Sierra Leone, likely to offer the type of low-risk, high-return scenarios that Tanzania, Uganda, and Zambia. Many of the most recently characterize those infrastructure projects currently success- created and planned SWFs with a domestic investment man- ful in attracting foreign SWF financing. date are in resource-rich countries. Furthermore, lack of government capacity for project se- lection, appraisal, design and implementation; weak gover- Why Are Governments in Resource-Rich nance and regulatory frameworks; and lack of coordination Countries Looking at SWFs to Finance among government entities, as well as political economy is- Domestic Infrastructure? sues, often compound the challenges faced by developing There appear to be a number of reasons why many govern- countries in planning and executing investment projects ments wish to increase the role of SWFs in financing domes- through traditional public expenditure channels, as well as in tic infrastructure. One is the decrease in traditional sources attracting private investors. Thus the question remains, of financing for infrastructure in developing countries since would an SWF authorized to spend domestically be better the recent global financial crisis. At the same time, infra- suited to face these types of challenges? structure needs in these countries remain high, so popular Central issues are the relationship of SWF financing to sentiment may push the government to spend part of its ac- the budget process and to procurement systems of sector min- cumulated financial wealth domestically. In addition, public istries, as well as the establishment of appropriate bench- investment in resource-rich countries often highlights sig- marks and safeguards to ensure the integrity of investment nificant management and governance challenges, including decisions. Depending on the size of domestic investment by low capacity, weak governance and regulatory frameworks, the SWF, consistency with macroeconomic stabilization poli- and lack of coordination among public entities (Dabla-Nor- cies would likely be an issue: the SWF could exacerbate mac- ris et al. 2011). Against this backdrop, some governments roeconomic and asset-price cycles by investing heavily in do- may see the SWF as a means to improve the quality of public mestic infrastructure when resource prices and the economy spending, and even to crowd in private investors to strength- are booming. en investment discipline. The SWF could also be used to bypass parliamentary The use of government revenue and export earnings scrutiny of spending, resulting in inefficient and fragmented from the ownership and taxation of natural resources pres- public investment programs. Adding the SWF to the list of ents a substantial macroeconomic and intertemporal chal- entities authorized to invest domestically could compound lenge. This revenue may be partly saved in an SWF to provide the risks of wasteful expenditure, budget fragmentation, po- benefits for future generations. It can be used to reduce pub- litical capture, and lack of coordination with fiscal policy, es- lic debt, smooth the effect of resource revenue volatility, or be pecially in low-capacity and low-governance environments. spent on various government objectives. However, the ability By receiving its funds from excess reserves—instead of raising of the economy to absorb spending delimits the size and effi- funds on financial markets as domestic development banks ciency of domestic investment by the government. To avoid do—the SWF’s operation would not be subject to similar mar- waste and overheating of the economy, a large chunk of this ket discipline. The volatility of excess reserves from extrac- revenue is typically invested externally. tives compared to the long-term nature of infrastructure proj- ect commitments, and the somewhat limited possibility of using real options to increase portfolio flexibility, may also af- Why Is Domestic Investment by an SWF a fect the type of projects and investment mechanisms available Tricky Proposition? to the SWF. However, although these risks cannot be elimi- nated, several measures exist to mitigate them. Recent macroeconomic studies suggest that domestic invest- ment of excess reserves has the potential to raise economic What Institutional Measures Could Mitigate growth and diversify the economy away from nonrenewable the Risks? resources (Berg et al. 2012; Collier et al. 2009; van der Ploeg and Venables 2010). However, the use of SWFs to finance do- Competitive investments mestic infrastructure holds significant challenges. Allocations for domestic investment should not be fixed at a Existing literature points to the difficulty of separating certain portfolio share, but rather determined on the basis of SWFs’ domestic investment decisions from political interfer- competition with foreign assets. In periods of low domestic ence and elite capture. As stated, transparent policy of invest- returns, or when there are indications of asset bubbles, invest- ing on a commercial basis, with allowance for lower returns as ments would be channelled abroad. If the investment project a trade-off for public utility, is unlikely to eliminate such pres- has a clearly defined development objective, it would still be sures: social returns are often difficult to measure, and domes- benchmarked against the financial return on foreign assets, 2 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise but allowances could be made for a limited mark-down from ii. Professional staff. To operate as an expert investor, the the benchmark rate, as discussed in greater detail below. In- SWF needs to be staffed with well-trained professionals, vestments that do not abide by these criteria cannot be ex- just like any financial investor in the private sector. pected to yield a competitive return. This is, for example, the iii. Transparent reporting. Consistent with good practice, case of public schools, which should be financed by the na- SWFs permitted or mandated to invest domestically tional budget. should issue publicly available reports covering their ac- Pooled investments tivities, assets, and returns. If the portfolio is partially Investing with private investors, pooling with other SWFs, market based and partially invested in projects with be- and cofinancing with international financial institutions low-market returns, these should be reported separately. could be used by the SWF to reduce risk, bring in additional iv. Independent audit. An internal audit should be reported expertise, and enhance the credibility of the investment deci- directly to the board, while an external audit ought to be sion. The Nigeria Infrastructure Fund, having signed coopera- conducted by an internationally reputable firm that is tion agreements with General Electric, the Africa Finance independent of the state. Corporation and the International Financial Corporation, What Should the SWF Invest In and How? exemplifies this approach (Rice and Blas 2013). Co-investing is standard practice used by institutional investors to manage Projects are not created equal. Although a well-governed SWF risk, and crowd in strategic partners. Where there is a risk that with a sound mandate and professional management and investment decisions may be affected by political and lobby- staffing can possibly improve the quality of the public invest- ing pressure, limiting the role of the SWF to that of a minority ment program, its mandate should not duplicate that of other investor can help to strengthen the integrity of the investment financial institutions (for example, the budget or any domes- process. Taking advantage of its long-term horizon, the SWF tic development bank). In other words, the scope of domestic could offer a range of instruments to share risk and make SWF investments should be limited to those appropriate for a commercially attractive projects viable for the market. wealth fund. Through innovative public-private partnership arrange- Public investments can be evaluated from two perspec- ments, the SWF may accept a somewhat lower return on mar- tives: (i) their financial or private returns and (ii) their broad- ginally commercial projects with large social benefits, thereby er economic and social returns. The latter include positive or making the projects attractive for the private sector. negative externalities for the wider economy and society that Strong corporate governance can cause the social rate of return to be higher or lower than Sound corporate governance is a prerequisite for effective and the financial rate of return. For example, an infrastructure sustainable performance. It provides incentives for manage- project might have positive economic or social externalities ment to take actions that lead to the achievement of the share- that are not fully captured by its financial return. holder’s objective, and it facilitates performance monitoring Figure 1 illustrates the universe of investment possibili- by shareholders (Canada 2005). There is a large body of ties at the country level along two dimensions: the private and knowledge on effective external (relationship between the the social return.2 Investments can be classified according to SWF and the state) and internal (composition and function- whether their financial returns pass a “market test,” R. Those ing of the board of directors or trustees and the SWF’s man- that pass lie in segments C and D, and will be attractive to agement processes) governance arrangements.1 The following private investors, including foreign SWFs; those that fail lie aspects of governance are particularly relevant to SWFs autho- below. Figure 1 also shows whether the economic returns ex- rized or mandated to invest domestically: ceed an acceptable threshold, E. Those that fail this test are to i. Independent board. Government officials often serve as the left of point E. The diagonal line separates investment op- board members for state-owned entities. Combining portunities that offer positive externalities (to the right) from ownership and supervisory roles presents conflicts of in- those that provide only private benefits and that would not terest that may undermine the integrity of both func- qualify for public investment. tions and expose decision making to political capture. To Investments in segment A should never be undertaken, ensure an adequate level of professionalism and indepen- since they provide neither adequate financial or economic re- dence of the board (both actual and perceived), all board turns. Investments in segment B provide high social or eco- members should meet specific skills and experience re- nomic returns, but low financial returns, such as rural roads, quirements. Nomination committees comprising indi- schools, or health facilities. These investments should be viduals deemed to be independent and objective can help funded through the normal budget process. If the SWF’s eli- ensure a politically independent selection process, al- gible investments should include this category of spending, though it is difficult to expect perfect independence any practical distinction between the fund and the budget when the owner is the state. would vanish, the SWF would lose accountability, and its vul- 3 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise Figure 1. Investments by Financial and Social/Economic Returns this end, the capital objectives used by devel- threshold for social & opment banks and other financial institutions economic returns to ensure financial soundness of investments with a developmental purpose provide inter- esting examples. These include (i) achieving a minimal return that exceeds inflation (Finan- ciera Rural of Mexico and Credit Bank of Tur- higher line D D key), (ii) generating a rate of return that equals financial returns if credit C constrained or exceeds the government’s long-term bor- market test line rowing costs (Business Development Bank of Rc Rc (where private & C+ foreign assets Canada), and (iii) an explicit target return on R R C++ breakeven) capital, ranging from 7 to 11 percent annually A B (Development Bank of Samoa, EXIM Bank of A India, and Kommunalbanken of Norway). 45° The International Finance Corporation has E developed a new financial valuation tool, the social and economic returns Sustainability Program Quality Framework, Source: Authors’ illustration. which attempts to capture the full value of sustainability/social programs. The tool, nerability to political capture would increase. On the other which is an attempt to remove the subjectivity of ratings, is hand, if an SWF were to invest in segments C or D according currently being piloted. to purely commercial market principles, it could run the risk While this methodology holds promise for SWFs, a pos- of simply displacing private investors. sible alternative approach could be for the government to set To summarize, as a wealth fund, an SWF should not in- the overall target return on investment for the SWF’s portfo- vest in projects that are justified primarily by their econom- lio and the threshold minimum rate of return for all invest- ic or social externalities. Such investments should be fund- ments (for example, the government’s average long-term real ed through the normal budget process, which should also borrowing rate on commercial loans). The SWF would then make provision for the future recurrent costs necessary for be free to decide on the composition of its investment portfo- operations and maintenance. SWF investment not warrant- lio to maximize the overall rate of return, while guarding ed on commercial grounds would greatly complicate the ac- against investing in a project with expected negative returns. countability of the fund because its management could no For clear accountability, it is also important to separate the longer be benchmarked on financial returns. The SWF also below-market portion from the market-based portfolio. may not be accountable for the wider social and economic Conclusion impacts of investments, which may depend on factors out- side its control. Though not entirely new, SWFs permitted or mandated to The SWF should therefore screen domestic investment invest domestically are emerging on a wider scale. However, proposals primarily according to their financial return and they have not been systematically assessed, therefore there is seek development opportunities with market or close-to-mar- much to learn about their processes and activities. More re- ket financial returns in areas where it can crowd in, rather search is needed to better understand their operations and than displace, private investors. In some circumstances, the potential role for financing in developing countries. fund may accept a somewhat below-market return on domes- Since SWFs permitted or mandated to invest domesti- tic investments with large economic benefits. For SWF invest- cally combine features of traditional SWFs and development ments not fully justified on commercial grounds, it is essen- banks, they can draw on good practice examples from both tial to have a clear and transparent process for benchmarking types of institutions. Establishing rules on the type (for ex- financial returns and for trading off financial and nonfinan- ample, commercial and/or quasi-commercial investment) cial goals. and modalities (for example, no controlling stakes, leverag- Even if the theory is clear, its application is not without ing private investment) is one way to ensure separation be- challenges. Identifying what constitutes an “acceptable quasi- tween the activities of the SWF and those of other govern- market return” or “home bias” involves country-specific and ment institutions with investment mandates, such as the project-specific considerations and poses the risk of reducing budget, the national development bank, the investment au- public accountability because the measurement of economic thority, and state-owned enterprises. The critical issue re- benefits is more ambiguous than that of financial returns. To mains that of limiting the SWF’s investment scope to those 4 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise appropriate for a wealth fund. If investments that generate References quasi-market returns are permitted, the size of the home Balding, Christopher. 2008. “A Portfolio Analysis of Sovereign bias should be clearly stipulated and these investments Wealth Funds.” HSBC School of Business; ESADE University should be reported separately. Faculties, ESADEgeo. The overall objective is to create a system of checks and Berg, Andrew, Rafael Portillo, Shu-Chun S. Yang, and Luis-Felipe balances to help ensure that the SWF does not undermine Zanna. 2012. “Public Investment in Resource-Abundant Devel- macroeconomic management or become a vehicle for politi- oping Countries.” IMF Working Paper 12/274. cally driven “investments.” The difficult environments in Canada, Office of the Auditor General. 2005. Report of the Auditor General, chapter 7, http://www.oag-bvg.gc.ca/internet/English/ which some SWFs are being established suggest that these parl_oag_200502_07_e_14927.html). will often be major concerns. Only if the SWF is allowed to Collier, Paul, Frederick van der Ploeg, Michael Spence, and Anthony operate as a professional expert investor can it strengthen the J. Venables. 2009. “Managing Resource Revenues in Developing management of the public investment program and contrib- Economies.” OxCarre Research Paper 15, Oxford. ute to building national wealth. Dabla-Norris, Era, Jim Brumby, Annette Kyobe, Zac Mills, and Chris Papageorgiou. 2011. “Investing in Public Investment: An Acknowledgment Index of Public Investment Efficiency.” IMF Working Paper 11/37. This note draws from Gelb, Tordo, and Halland, "Mobilizing Gelb, Alan, Silvana Tordo, Håvard Halland. Forthcoming. “Mobi- Sovereign Wealth Funds for Long-Term Development Fi- lizing Sovereign Wealth Funds for Long-Term Development nance: Opportunity and Risks" (forthcoming). The authors Finance: Opportunities and Risks.” Economic Policy Working are grateful to Ekaterina Gratcheva, Christian Mulder, Grego- Paper, World Bank, Washington, DC. ry Smith, and Noora Arfaa for their insightful comments and IMF (International Monetary Fund). 2013. “Revised Guidelines contribution to the background paper. for Foreign Exchange Reserve Management.” Washington, DC. http://www.imf.org/external/np/pp/eng/2013/020113.pdf. About the Authors IWG (International Working Group of Sovereign Wealth Funds). 2008. The Santiago Principles for the Operations of SWFs. http:// Alan Gelb is a Senior Fellow at the Center of Global Develop- www.iwg-swf.org/pubs/gapplist.htm. ment in Washington, DC. Silvana Tordo is Lead Energy Econo- Monk, Ashby. 2013. “The Rise of Sovereign Development Funds.” mist for the Sustainable Energy Department, Extractive Indus- Institutional Investor. http://www.institutionalinvestor.com/ tries. Håvard Halland is a Natural Resource Economist for the blogarticle/3189172/Blog/The-Rise-of-Sovereign-Development- Funds.html. Poverty Reduction and Economic Management (PREM) Network OECD (Organisation for Economic Co-operation and Develop- of the World Bank. ment). 2004. Principles of Corporate Governance. Paris. Notes ———. 2006. Guidelines on Corporate Governance of State-Owned Enterprises. Paris. 1. These include the Santiago Principles for the Operations of Rice, Xan, and Javier Blas. 2013. “Nigeria SWF Makes Maiden In- SWFs (IWG 2008), the Revised Guidelines for Foreign Exchange vestment.” Financial Times, September 16, http://www.ft.com/ Reserve Management (IMF 2013), as well as general principles intl/cms/s/0/d0751b26-1ee8-11e3-b80b-00144feab7de. html#axzz2qII7WFoF. of good corporate governance practice, such as the Principles Truman, Edward M. 2011. “Sovereign Wealth Funds: Is Asia Dif- of Corporate Governance (OECD 2004) and the Guidelines on ferent?” Working Paper 11-12, Peterson Institute, Washington Corporate Governance of State-Owned Enterprises (OECD DC. 2006). Van der Ploeg, Frederick, and Anthony J. Venables. 2010. “Absorb- 2. This figure is not meant to represent the strategic asset al- ing a Windfall of Foreign Exchange: Dutch Disease Dynamics.” location model of a single entity. OxCarre Research Paper 52, Oxford. 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. 5 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK    www.worldbank.org/economicpremise