Christian Noyer: Foreign reserve accumulation – some systemic
Speech by Mr Christian Noyer, Governor of the Bank of France, at the Salzburg Global
Seminar, Salzburg, 1 October 2007.
"The management of international reserves
High current account surpluses mechanically reflect an excess of domestic (private or public) savings, which must be invested abroad. But in East Asia and commodity-exporting countries, these accumulated excess savings appear to be largely held by sovereigns. In addition to foreign exchange reserves, a growing part of these foreign assets is now invested through sovereign wealth funds (SWFs).
For the time being, the assets managed by SWFs (around USD 2.5 trillion) are relatively small compared with the global capitalisation of bond and equity markets – about USD 100 trillion – or even compared with the holdings of the private asset management industry (pension, insurance and mutual funds), estimated at around USD 55 trillion. However, according to some estimates, SWF assets could more than treble over the next 5 years and reach around USD 8.5 trillion in 2012.
All SWFs share a common purpose which is the transfer of wealth across time. The objectives of central banks are different, namely to foster price stability and financial stability. Besides, central banks may be more sensitive to headline risk, and exposure to large losses could damage their credibility. Creating an entity separate from the central bank, an SWF, in order to manage risky assets has often been seen as a good governance structure. Such a structure is also expected to set up a Chinese wall between central banks and SWFs, eliminating the risk of trading based on insider policy information. The effectiveness of such segregation may yet be challenged, but more importantly, it may not completely prevent conflicts of interest.
A distinction probably has to be made between commodity and non-commodity SWFs. In oil-producing economies, foreign asset accumulation stems from oil royalties reflected in large government budget surpluses, so that their management by public entities is a natural outcome. The funding of commodity SWFs stems from a domestic resource that is most of the time owned, exploited or taxed by the government: in this respect it is genuine sovereign wealth, and the result of public savings. By contrast, foreign reserves and SWF accumulation that is not commodity related might be seen as primarily a diversion of excess private savings.
The set-up of stabilisation funds by oil-exporting c ountries was first motivated by the desire to smooth oil revenues, which are highly volatile, with a view to maximizing the overall rent from oil extraction. To the extent that commodities are non-renewable resources, standard economic theory suggests that part of extraction revenues should be saved in order to smooth the nation’s inter-temporal consumption, very much like an individual saves over the lifecycle for his retirement, but also possibly in order to leave a bequest to his offspring.
Commodity SWFs were indeed equally motivated by an attempt to optimise the inter- temporal benefits of natural resources beyond the exhaustion horizon, as well as on the grounds of fairness to future generations. As such, oil funds are far from being a new phenomenon: the earliest funds were thus created in the aftermath of the first two oil shocks. The more recent surge in oil prices has naturally spurred the set-up of new funds in countries where new extraction capacities have emerged as profitable.
Non-commodity SWFs are mainly born out of foreign exchange interventions, and can be regarded as “spillovers” from international reserves that can no longer be properly managed within the standard framework of official re serve assets. One of the questions to be discussed is whether they correspond to the same rationale as commodity funds, and to what extent they can be considered to have the same sound basis.
FX reserves are typically invested in safe and liquid assets such as T-bills, T-bonds or short-term collateralised deposits. The emphasis on liquidity risk control in the prudent management of reserves logically results from their possible use in foreign exchange interventions.
But SWFs are clearly designed to invest in less-liquid and riskier assets that provide higher returns than the typical assets held in official reserves. Whereas the objective of reasonable returns was typically mentioned last, as “subject to liquidity and other risk constraints”, in the IMF Guidelines on reserve management, excess returns clearly feature as a primary
objective of SWFs. A central reason for this reordering of objectives is of course the much longer-term horizon of these funds, which entails greater tolerance of short-term fluctuations in returns. Indeed, the development of SWFs can in some cases to be related to the increasing reliance on fully-funded government pension schemes, which often share a similar approach to asset allocation.
For instance, Norway’s Government Pension Fund – Global, the continuation of its Petroleum Fund, is managed with respect to a strategic benchmark that holds 40% of its assets in equities, 50% of which are outside Europe.
In the case of several East Asian economies, the transfers from official reserves to SWFs therefore entail a loss of the liquidity services from holding reserve assets. But to the extent that expected returns are higher, one benefit of investing in riskier assets is a reduction in the fiscal cost of sterilisation.
The strategic asset allocation of SWFs could in principle be designed in relation to the specific risk exposures of the country. Emerging market economies are characterised by arange of vulnerabilities to global macroeconomic shocks such as sharp changes in oil, metals or agricultural commodities prices, as well as the business cycles fluctuations of developed economies. For instance, it seems intuitive that the optimal asset allocation of oil-producing countries should be diversified into assets uncorrelated with oil prices. By contrast, most Asian countries are commodity importers and could hedge their risks with long positions on commodities futures and other assets that are highly correlated with commodities prices. SWFs can thus potentially provide macro-hedging services to their countries. The task might however be easier for commodity exporters: their macroeconomic risks are clear, highly concentrated and hedging products are readily available on financial markets.
Impact on asset prices
Under what conditions can sovereign asset management go hand in hand with orderly
functioning of the financial markets?
Large foreign reserve holdings are already seen to have an impact on asset prices. The strong preference for safety and liquidity in official reserves, combined with the fast increase in reserve assets, has often been given as an explanation for what Alan Greenspan first called a conundrum, namely the persistently low level of long-term interest rates. For instance, Warnock and Warnock estimated that, under the hypothesis of “no foreign official flows into U.S. government bonds in 2005”, the 10-year Treasury yield would have been 90 basis points higher. Several alternative estimates exist and there are significant differences between them. But they consistently conclude at least some impact on long-term interest rates.
As SWFs hold portfolios with higher risk/return profiles than the usual FX reserve management funds, many market economists predict that their development will result in additional demand for high-risk asset classes like equities and lower demand for low-risk assets classes like short-term government securities. This could in theory lead to an increase in bond yields (partly reversing the prior impact of official reserve accumulation) as well as a decrease in the equity risk premium. Estimates of these effects must however be interpreted with caution: they rely on fragile assumptions, amounting to an equivalency between the development of risky sovereign holdings and a lasting decline in global risk aversion. Before jumping to such a conclusion, one should keep in mind that if private savings had not been intermediated in public balance sheets, they might as well have been invested directly in risky assets.
Yet the impact of SWF investments on some small and illiquid financial markets, such as gold, commodities and real estate might be high. To the extent that SWFs attempt to hedge macroeconomic risks for their country, difficulties could arise insofar as the best existing hedging instruments may be less liquid than bonds and equities. The risk of a potentially disruptive impact on asset price dynamics should then be acknowledged for large SWFs, as underlined by the 2007 IMF Global Financial Stability Report..."