Sovereign wealth funds (SWFs) are saving funds controlled by sovereign governments that hold and manage foreign assets. While not a new phenomenon, the recent activities and projected growth of SWFs have stirred debate about the extent to which their size may allow them to affect financial markets and their policies may be driven by non-economic considerations. As an increasing share of the foreign asset holdings of sovereign wealth funds shift from government debt obligations to private equities, concerns also have arisen about how institutions in the “investing” and “receiving” countries may need to adapt. Much discussion (Johnson 2007 and Economist 2007) has been devoted to the need for individual SWFs to be more transparent about their investment approach, by providing more information on the type and amounts of assets they hold, and about their governance structure, by clarifying how decisions are made and monitored.
Sovereign wealth funds, unlike monetary authorities holding official foreign reserves, typically seek to diversify foreign exchange assets and earn a higher return by investing in a broader range of asset classes. Sovereign wealth funds, unlike mutual and insurance funds, generally have no specific liabilities to be paid to shareholders or policyholders. Sovereign wealth funds similarly differ from sovereign pension funds in that the latter have explicit liabilities, such as worker pensions. For this reason, SWFs typically have had less incentive to be transparent about their investment and management practices. However, as SWFs invest more of their assets in private financial markets, greater concern has arisen as to the extent to which they should follow the practices of private institutional investors and pension funds in industrial countries.
In a recent paper (Aizenman and Glick, 2008), we discuss the forces leading to the growth of sovereign wealth funds and provide statistical analysis supporting stylised facts about their determinants and behaviour. Countries with SWFs tend to be fairly evenly distributed across income levels – there are high-, medium-, and low-income countries with SWFs. Applying probit regressions of the incidence of SWFs, we confirm that countries that run larger current account surpluses or that specialise in fuel exports are more likely to establish sovereign wealth funds. Not surprisingly, the ratio of foreign reserves to GDP is relatively high in many countries with SWFs, particularly in countries that have recently established such funds.
Sovereign wealth funds and governance
Sovereign wealth funds differ in their strategies for investing abroad and in the information they provide about their activities. The countries of origin of SWFs also vary in governance standards at the national level. How does the governance and transparency of individual SWFS compare with norms of behaviour in home and foreign countries? To answer this question we use the Truman (2008) governance scores for individual sovereign wealth funds and the Kaufmann, Kraay, and Mastruzzi (2007) measures of the quality of national corporate governance.
Figure 1 compares the Truman governance scores for “older” oil-export based SWFs (United Arab Emirates, Qatar, Brunei, and Oman)1, Norway’s SWF, other SWFS, and sovereign pension funds of selected industrial countries. It shows that the older oil-exporting countries have lower Truman fund governance scores, primarily because of limited transparency and accountability, in comparison to other SWF countries. All are well below the standards of sovereign pension funds in industrial countries as well as Norway’s SWF.
Figure 2 displays the average national governance measures for our SWF country groupings. It shows that the older oil-exporting countries generally have better governance levels than those of other SWFS (except Norway, of course), but lower democracy levels. These figures illustrate how the practices of many existing sovereign wealth funds, particularly those originating in less democratic countries, differ from the practices of industrial country pension funds as well as Norway's government fund. Clearly, there is still a great difference between the governance standards of the economies in which SWFs have been established and the governance standards of the industrial economies in which they are seeking to invest.
Figure 3 gives a scatter plot of the Kaufmann, Kraay, and Mastruzzi voice and accountability governance subindex, a measure of democracy, vs. Truman’s total SWF score. It suggests a positive correlation between the two measures. That is, countries characterised by greater democracy also tend to have SWFs displaying better overall governance.
The figure also shows that the oil-producing countries of the United Arab Emirates, Qatar, Brunei, and Oman (UAE, QAT, OMN, BRN, in the lower left quadrant) have relatively low democracy levels as well as SWFs with low Truman fund governance scores. In contrast, the SWFs of the newer fuel-producing countries – Russia, Kazakhstan, Azerbaijan, and Timor-Leste (RUS, KAZ, AZE, TMP, in the lower right quadrant) – who also have low democracy scores, have higher Truman fund scores, i.e. they are more transparent than the sovereign funds in the older oil-producing countries. Why might this be so? One possible explanation is that countries that have only recently begun to develop their fuel resources have a greater incentive to foster more global integration by establishing institutions, such as SWFs, with more transparency and accountability. It is easier to change the level of transparency of a fund than to change a country’s political system.
Sovereign wealth funds and foreign reserves
As countries choose to entrust more of their sovereign wealth to investment funds, official reserve accumulation is expected to slow, with funds likely shifting away from reserves held by the central bank. To assess this possibility, Figure 4 depicts the asset holdings of SWFs and the foreign reserve holdings of central banks, expressed as shares of GDP (at the end of 2007 and 2006, respectively), ranked by the fund’s age (years since establishment). The older SWFs not only have relatively high SWF-asset-to-GDP ratios, they also have relatively low foreign-reserve-to-GDP ratios compared to most of the newer SWFs. This is consistent with the view that over time countries may transfer a greater share of the public sector’s foreign assets from official reserves into SWFs.
However, a panel regression of the determinants of foreign reserves relative to GDP over the period 1985-2006 does not support the view that the establishment of a SWF eventually leads to lower official reserve levels. One possible reason is that most of the SWFs in the sample have been established relatively recently, leaving insufficient time for effects on the level of central bank reserve holdings to occur.
In our paper, we also present a model with which we compare the optimal degree of diversification into safe foreign reserve assets and other higher-yielding, but risky, foreign assets by a central bank versus that of a sovereign wealth fund. We show that if the central bank manages its foreign assets with the objective of reducing the probability of sudden stops, it will place a high weight on the downside risk of holding risky assets abroad and will tend to hold primarily safe foreign assets. In contrast, if the sovereign wealth fund, acting on behalf of the Treasury, maximises the expected utility of a representative domestic agent, it will opt for relatively greater holding of more risky foreign assets. We also show that, as a country’s overall foreign asset base increases, the opportunity cost associated with the limited portfolio diversification of the central bank induces authorities to establish a wealth fund in pursuit of higher returns.
The present global financial crisis illustrates the importance of the precautionary purpose for holding international reserves by central banks. While the recent drop in global commodity prices and equity returns may have reduced the relative appeal of sovereign wealth funds, a resumption of global growth may restore their attractiveness. However, if the “great moderation” period is indeed over and volatility in financial markets remains high, monetary authorities may place a high weight on holding more reserves as a means of minimising the expected costs of sudden stop crises.
Editors' note: The views expressed do not represent those of the Federal Reserve Bank of San Francisco or the Board of Governors of the Federal Reserve System.
Aizenman, Joshua and Reuven Glick (2008), “Sovereign Wealth Funds: Stylized Facts about their Determinants and Governance,” NBER Working Paper No. 14562.
Economist (2007), “Fear of Foreigners,” August 14 (online only).
Johnson, S. (2007), “The Rise of Sovereign Wealth Funds,” Finance and Development, 56-7.
Kaufmann, Daniel, Aart Kraay, and Massimo Mastruzzi (2007), “Governance Matters VI: Aggregate and Individual Governance Indicators, 1996–2006,” World Bank Working Paper WPS4280.
Truman, E. (2008), “A Blueprint for Sovereign Wealth Fund Best Practices,” Peterson Institute for International Economics, Policy Brief No. PB08-3, Washington DC.
1 There is no Truman data for Saudi Arabia.