Petrodollars, Portfolio Restructuring and Long-Term Interest Rates

Thomas Barnebeck Andersen, Sofie Nedergård Fick and Frank Øland Hansen, Financial Markets

INTRODUCTION AND SUMMARY

Global financial imbalances have widened significantly in recent years and the US current-account deficit, which was previously financed primarily by the Asian emerging market economies, is increasingly financed by the oil-exporting countries, cf. Table 1.^[1] Following the tripling of oil prices, from just over 20 dollars per barrel in 2002 to 72 dollars per barrel in August 2007, the oil-exporting countries now finance most of the US current-account deficit. Therefore, the savings and investment strategies of these countries have a major impact on global demand for government bonds – and thus on long-term interest rates.^[2]

The current rise is the third steep increase in oil prices. While the high oil prices during the oil crises of 1973-74 and 1979-80 were caused by sudden slumps in the supply of oil, today's high oil prices are driven mainly by increased demand. At the same time, the industrialised economies have become less oil-intensive^[3], and thus more resilient to oil-price rises. Consequently, the current oil-price rises have a more modest impact on inflation and economic growth in oil-importing countries than was the case in the 1970s.

During the two oil crises, international financial markets were less developed than today and the oil-exporting countries did not invest in securities to the same extent as they do now. Instead, much of the oil revenue was deposited with international banks, typically US banks. Most of the extraordinary revenue of the oil-exporting countries was, however, allocated to increasing imports, not least in response to rapidly rising public consumption and major infrastructure investments.

This time around, the oil-exporting countries have saved up more of the oil revenues and have invested most of them in safe assets, such as US government bonds. This has contributed to a downward pressure on long-term interest rates. In line with the accumulation of substantial wealth in recent years, the oil-exporting countries have also become more willing to assume risk. Consequently, an increasing proportion of the assets are managed by entities other than the central banks of the oil-exporting countries. These sovereign wealth funds first seriously hit the " radar screens" of international market participants around the turn of the year. Since they are not subject to the same restrictions as central banks, they are able to invest in riskier assets. Several other countries, including China, are currently setting up similar funds. Such changes to the investment strategies of countries with large current-account surpluses tend to reduce the demand for government bonds, which could exert upward pressure on long-term interest rates.

This article discusses ongoing changes to the investment strategies of the oil-exporting countries and the potential impact on long-term interest rates.

OIL PRICES AND OIL REVENUES

A steep increase in oil prices generates a similar increase in the export revenues of the oil-exporting countries. The higher oil revenues may be employed in one of two ways. Oil exporters may increase imports of goods and services, or acquire foreign financial assets, mainly in the form of deposits with foreign banks and investments in foreign stocks and bonds.

Since oil prices really took off in 2003, the oil-exporting countries have allocated just under half of the increase in export revenues to higher imports, i.e. consumption and domestic investments. The rest of the increase manifests itself as a current-account surplus, cf. Chart 1.^[4] The Chart clearly illustrates the close correlation between oil prices and current-account surpluses in the oil-exporting countries. It is a rule of thumb that for every 10-dollar increase in the price of oil, the oil-exporting countries typically generate a further 110 billion dollars of current-account surplus.

As a consequence of the rise in oil prices, the oil-exporting countries now boast current-account surpluses that exceed those of the Asian emerging market economies and Japan. In 2006, the oil-exporting countries had an estimated current-account surplus of 571 billion dollars, up from 88 billion dollars in 2002, cf. Table 1.

HOW DO THE OIL-EXPORTING COUNTRIES INVEST THEIR WEALTH?

The USA, with a deficit of 869 billion dollars in 2006, is the primary counterpart of the oil-exporting countries' current-account surplus of 571 billion dollars in 2006, cf. Table 1. Thus it is to be expected that the oil-exporting countries are re-investing a large portion of their surpluses directly in the USA. This is difficult to prove, however. The problem is that, with a few exceptions, the oil-exporting countries do not publish breakdowns of their foreign assets. Therefore, secondary data sources must be used, e.g. information provided by securities issuers. However, it is often not possible to identify the final buyer based on secondary data sources – one of the primary reasons being the growing complexity of international financial transactions involving international central securities depositories, offshore financial centres and hedge funds. For example, an oil exporter can invest funds in US government bonds using a London-based depository. The issuer, in this case the US Treasury, will record this transaction as a purchase from the UK.

Three secondary data sources are typically used. Firstly, the US Treasury Department reports on foreign purchases of US securities, broken down by different maturities. Secondly, the BIS publishes figures on banks' deposits from and lending to individual countries. Thirdly, Bloomberg publishes data on mergers and acquisitions. A comparison of the current-account figures of each oil-exporting country with these data sources provides an indication of where the oil money is going and how much of the total savings it is possible to account for.

Toloui (2007) has conducted this analysis and finds that for the period 2002-06 only about 40 per cent of the total savings of the oil exporters can be accounted for. Some 2/3 of this identifiable portion was invested in bank deposits and other safe assets, including government bonds, primarily denominated in dollars. However, the country data coverage is somewhat uneven: while most of the Russian surplus can be accounted for, it is possible to account for only 27 per cent of the surplus of the Middle Eastern countries. Russia is the driving force behind the observable demand for safe assets, with investments in safe assets amounting to about 80 per cent, while the Middle Eastern countries invested only about 50 per cent of their identifiable funds in safe assets.

Statistically, it is not possible to achieve a more complete picture of how the oil exporters have invested their current-account surpluses. However, economic theory can give an indication of how investment strategies may change as oil wealth accumulates. In the following section, we will explore how the risk appetite of the oil exporters may change over time – and the impact of this change on long-term interest rates. 

OIL SAVINGS AND LONG-TERM INTEREST RATES

A transfer of revenue from the oil-importing countries to the oil-exporting countries, as a result of higher oil prices, can impact long-term interest rates in two ways.

Firstly, the oil-exporting countries' high propensity to save will increase global savings. Other things being equal, this will exert downward pressure on global long-term interest rates.

Secondly, in addition to impacting the level of global savings, a transfer of revenue will determine which assets are in demand. If the countries have different preferences, a transfer of revenue from the oil-importing countries to the oil-exporting countries will influence the relative demand for asset types. If the oil-exporting countries are less risk averse than the oil-importing countries, higher oil prices will, other things being equal, dampen the demand for government bonds through this portfolio effect and exert upward pressure on interest rates.

Depending on which effect is dominant, the overall effect on long-term interest rates of a transfer of revenue from the oil-importing countries to the oil-exporting countries can be positive or negative. Traditionally, most attention has been given to the savings effect, but the portfolio effect is increasingly coming into focus. The weakening of the savings effect and the strengthening of the portfolio effect after a protracted period of high oil prices is entirely in accordance with economic theory.

Under the so-called permanent income hypothesis, the savings effect will be reduced as oil exporters begin to perceive the high level of oil prices as permanent^[5] – the reason being that the need to save, to achieve a steady development in consumption, is reduced if oil revenues are expected to remain high in the future.

The opposite applies to the portfolio effect, the significance of which is expected to increase as the assets of the oil-exporting countries grow. It seems fair to describe large investment funds as having a decreasing relative risk aversion, i.e. their risk aversion decreases as their assets grow.^[6] Under a portfolio allocation model, this assumption would mean that the willingness to hold a larger share of risky assets increases as assets grow. Consequently, the proportion of investments placed in government bonds falls as the assets of the oil-exporting countries grow. This could lead to reallocation of existing portfolios generated from oil revenues, and is probably what we have been witnessing since 2002. Examples of current portfolio restructuring in oil-exporting countries are provided in Box 1.

Long-term interest rates
In early 2002, the oil-exporting countries focused on safe dollar investments, such as US government bonds. This approach, combined with a positive savings effect, has probably helped to push down long-term interest rates. In line with the continued asset accumulation resulting from higher oil prices, the investment strategies of the oil-exporting countries have gradually become less risk averse and the countries have, little by little, placed a larger proportion of their investments in risk-related assets. This substitution towards riskier assets has probably exerted gradual upward pressure on interest rates. The dynamics described are consistent with the actual development in interest rates and oil prices over the period 2002-07, cf. Chart 2.

Caution should be exercised in interpreting the direction of causation based on simple time series, such as oil prices and interest rates in Chart 2. It could be just a statistical coincidence. Firstly, real interest rates exhibit a high degree of sluggishness. Secondly, no adjustment has been made for other factors that affect both variables. Oil prices have increased in parallel with the expansion in the global economy. It may therefore be difficult to determine whether a correlation between oil prices and interest rates should be attributed to the oil-price effect on interest rates or whether both factors have been driven by the expansion in the global economy, including the US economy.^[7] However, the correlation described between oil prices and long-term US interest rates also holds true if adjusted for cyclical effects through regression analysis. Given the short time span, the sluggishness of interest rates and the trend in oil-price movements – i.e. the tendency of oil prices to rise throughout the period under review – major statistical problems are associated with regression analysis.

Effect of declining risk aversion
A calibration exercise is an alternative way to quantify the effect of portfolio restructuring. The idea is to simulate a relevant economic model by choosing realistic parameter values that will help to ensure that the predictions of the model are, to all intents and purposes, in line with reality. Once the model has been calibrated, individual parameters can be adjusted, e.g. risk aversion, in order to study their impact on the relevant variable – in this case long-term interest rates.

The so-called CCAPM, outlined in Box 2, is an appropriate model framework for this purpose. The model is a standard model framework in finance theory. As opposed to relative pricing models, e.g. the classic Black-Scholes model for pricing of options where prices are fixed relative to an underlying asset, CCAPM is an absolute pricing model where prices are based on the actual exposure of the asset to fundamental macroeconomic risks.

Using a CCAPM framework, Jen and Miles (2007) have studied the correlation between risk aversion and long-term US interest rates.^[8] They have found that for realistic parameter values, 10-year US interest rates may increase by about 40 basis points in the coming years. It must be stressed that the model outcome is subject to uncertainty, and that a number of reservations should be made. The outcome shows that – within the framework of a standard economic model – increased risk tolerance may, other things being equal, lead to higher long-term interest rates, and the impact may be significant. 

CONCLUSION AND PERSPECTIVE

In this article, we have attempted to demonstrate how the growing assets of the oil-exporting countries may exert upward pressure on long-term interest rates. The article has not taken the Asian central banks into consideration. It should be noted, however, that the pressure on long-term interest rates could be strengthened by likely portfolio restructuring in Asia, notably in China.

China holds the world's largest foreign-exchange reserve totalling approximately 1,200 billion dollars. So far, the country has placed the bulk of these assets in US government bonds. Therefore, China attracted considerable international attention last March when announcing that it would set up a new investment fund, State Investment Company, modelled on the Singaporean investment fund Temasek. This decision reflects a wish for more diversification and a higher rate of return. Until now, China has announced that about 200 billion dollars will be actively invested, equivalent to roughly one sixth of the total foreign-exchange reserve.

The first visible evidence of China's new policy line emerged on 20 May 2007 with the announcement that China's State Investment Company would invest 3 billion dollars in Blackstone, the second-largest US private-equity fund.^[9] It is important to bear in mind, however, that 3 billion dollars is equivalent to just a few days of reserve accumulation in China, so China's foreign-exchange policy has not changed dramatically for the time being.

Finally, the emergence of large sovereign wealth funds could give rise to a number of problems for the international financial markets, cf. Lowery (2007). One concern is that lack of transparency could lead to increased volatility in the financial markets as different market participants act on their expectations of the contemplated actions of these funds. Another concern is that increasing financial protectionism could result from foreign sovereign wealth funds gaining control of sensitive domestic enterprises.

LITERATURE

Barro, R. (2006), Rare Disasters and Asset Markets in the Twentieth Century, Quarterly Journal of Economics, August

Barsky, R. and L. Kilian (2004), Oil and the Macroeconomy since the 1970s, Journal of Economic Perspectives, 18(4), pp. 115-134

Cochrane, J. (2005), Asset Pricing, Princeton University Press

Elton, E., M. Griber, S. Brown, and W. Goetzmann (2007), Modern Portfolio Theory and Investment Analysis, Wiley

Higgins, M., et al. (2006), Recycling Petrodollars, Federal Reserve Bank of New York, Current Issues in Economics and Finance, Volume 12, No. 9, December

IMF (2005), Will the Oil Market Continue to be Tight? World Economic Outlook, April

IMF (2006a), Statistical Appendix, World Economic Outlook, September

IMF (2006b), Oil Prices and Global Imbalances, World Economic Outlook, April

IMF (2006c), Regional Economic Outlook, Middle East and Central Asia, September.

Iversen, P. F. (2006), The USA's External Imbalance in a Financial Perspective, Danmarks Nationalbank, Monetary Review, 4th Quarter

Jen, S. and D. Miles (2007), Sovereign Wealth Funds and Bond and Equity Prices, Morgan Stanley Research Europe

Lowery, C. (2007), Remarks by Acting Under Secretary for International Affairs Clay Lowery on Sovereign Wealth Funds and the International Financial System, www.treas.gov/press/releases/hp471.htm

Toloui, R. (2007), Petrodollar, Asset Prices, and the Global Financial System, PIMCO Capital Perspectives