Abstract
The September 2015 issue of the IMF Research Bulletin covers a range of research topics and announcements of interest to those following finance and economics.
This article explores strategies for managing revenue from natural resources, focusing on the balance between domestic and foreign asset accumulation. It suggests that domestic asset accumulation is the priority, while there are three motives for accumulating foreign assets: inter-generational transfer, parking funds, and stabilization. The paper argues that the first of these is inappropriate for low income countries. The second is required if it is difficult to absorb extra spending in the domestic economy and takes time to build up domestic investment. The third is important, and depends on the extent to which the economy has other ways of adjusting to shocks.
The recent commodity super-cycle saw oil prices rise from below US$30 per barrel in 2003 to over US$100 per barrel in 2011, before falling to US$50 per barrel in 2014. Over the course of this cycle, global resource rents nearly tripled, from $7.6 trillion in 2000 to over $21 trillion in 2008 (constant 2005 US$, World Bank WDI). In some countries these revenues went straight in to current spending. In other countries natural resource funds were established and some of the revenues were placed offshore. However, there is little evidence that developing countries used revenues to make the domestic investments necessary for sustained growth in non-resource sectors of the economy. As this super-cycle comes to an end, now is an ideal time to evaluate resource policy and prepare for the future.
This research summary draws on work undertaken by the Oxford Centre for the Analysis of Resource Rich Economies (Oxcarre), in particular Venables and Wills 2015, which investigates the economic principles that should underpin use of resource revenues, especially in developing economies. We look first at the trade-off between using resource revenues for current spending (consumption) or for building up assets. We then turn to the question of what assets should be accumulated, focusing on the choice between capital assets in the domestic economy (human as well as physical) and foreign assets (e.g., held in a sovereign wealth fund). We argue that economic principles call for a high proportion of resource revenues to be used for building up assets and, in developing economies, these should be principally domestic capital.
Current Consumption and Asset Accumulation
What proportion of resource revenues should be saved (i.e., used to accumulate assets, real or financial, in the domestic economy or abroad) and how much should be used to finance current consumption? The benchmark answer to this is that spending from a temporary windfall should be smoothed through time according to the permanent income hypothesis (PIH), which says that the annuity value of the windfall should be consumed and the rest saved—with this savings producing a stock of assets to permanently finance the higher consumption. This is easily stated, but less easily applied and adapted to the circumstances of a developing economy. A multitude of issues arise, and we discuss just two.
First, how does the permanent income hypothesis translate into a simple rule for the proportion of revenue that should be saved? Intuitively, the shorter the duration of the expected revenue flow, the higher the proportion of revenue saved (if a one-day windfall were to finance a permanent increase in consumption, it would all need to be saved). For example, if resource revenue is expected to be a step function (a constant flow dropping to zero at date of exhaustion) and the interest rate is 4 percent, then 20 percent of resource revenue should be saved when exhaustion is 40 years away, rising to 45 percent at 20 years and 67 percent when exhaustion is 10 years away. If the expected decline in revenue is less abrupt, the numbers differ; but a faster rate of decline raises the required saving rate. Notice that, while this suggests high savings, the recommendation of 100 percent saving (sometimes referred to as the “bird-in-hand” rule) follows only if policymakers are so risk averse that they expect future revenues to be zero (exhaustion is imminent).
Second, the permanent income hypothesis needs modification for a developing economy in which current income is low and relatively rapid income growth is expected in the future. The modification is that current poverty makes it desirable to have a somewhat lower savings rate; essentially, it is not efficient to use the revenue to fund a permanent income increase that gives as much to future (and richer) generations as to the current (relatively poor) citizens. Formal analysis of this is in van der Ploeg and Venables (2011); the argument is that, in a capital-scarce economy, saving from resource revenue will bring down the rate of return on capital, flattening the efficient consumption path (the Euler equation) and implying a relatively large initial increase in consumption.
Applying these principles to a particular case is, of course, country and context specific, but some general messages come through. The rate of saving from resource revenues should be high, should increase as the resource stock is depleted, but should not be so high as to forego all short-run consumption benefits.
Domestic Capital and Foreign Assets
An extreme version of the PIH suggests that all saving should be directed into foreign assets, rather than be used to build up domestic capital. The argument is that the capital stock in the domestic economy earns the world rate of return on capital. Investing more would reduce the return below the world rate, and thereby be inefficient. It would be better to invest abroad and earn the world rate. This argument is sound for Norway and other capital abundant economies but is inappropriate for developing countries that are short of capital of all sorts—human, physical, and public. This is particularly the case in so far as resource revenues accrue to public funds on which there is a premium due to weak fiscal capacity. The returns gained from investing revenues in the domestic economy can be used to build capacity and to fund projects, such as infrastructure, which may in turn increase the level of private investment in the economy. The appropriate rate of return is then the full social rate of return, i.e., the direct benefits of the project plus the social value of induced effects in the private sector.
These arguments point to using resource revenues for domestic investment. What is the case for putting a fraction of them in foreign assets, through a sovereign wealth fund? There are two important arguments, which we refer to as “parking” and “stabilization.”
The parking argument turns on the ability of the domestic economy to make productive use of an increase in investment. Ramping up investment sharply creates a risk that projects undertaken will be of poor quality and low return. There may be bottlenecks, particularly in the supply of non-traded goods necessary for investment such as the construction sector (for physical capital), or the availability of skilled labor (including government capacity to implement projects and the supply of teachers needed for human capital investment). An increase in spending will then bid up prices and yield poor value. At a wider macro-level, the extra spending may lead to inflation and create the risk of boom and bust. There are two responses. One response is to plan ahead, anticipating bottlenecks and phasing investments appropriately—“investing-in-investing” in Collier’s (2010) terminology. The other response is to establish an offshore “parking fund” where revenues are accumulated until they can be invested productively at home. In summary, government needs to design an efficient investment program. The timing of this will, quite generally, not coincide with the timing of resource revenue receipts. Funds need to be held offshore and drawn down to finance domestic investment whenever it is efficient.
The parking fund smoothes anticipated gaps between resource revenues and domestic spending, but there will also be unanticipated revenue shocks driven by the volatility of resource prices. These shocks create an argument for some sort of insurance strategy. A stabilisation fund can play this role, as funds are deposited when prices are particularly high and drawn down when prices are low. What are the economic principles that should govern this and other responses to resource price volatility?
One possibility is that countries insure themselves against commodity price fluctuation, passing the risk to other economic agents. Contractual terms with foreign investors in the resource sector do this to a limited extent. Countries can also engage in hedging strategies, as practiced by Mexico, which purchases put-options to lock in the price of some of its oil sales up to a year ahead. These provide considerable insurance, but they incur transaction costs and only offer a relatively short period of protection.
Absent this insurance, fluctuations in resource revenues will impact countries’ expected wealth, and policy should ensure that these impacts do not have disruptive consequences arising from a sharp fall in foreign exchange receipts and/or government revenues. Access to international capital markets is, in principle, one way to manage this—borrowing when revenues fall. However, this option may be extremely expensive or simply unavailable if, for example, revenues fall during a time of global economic crisis. The alternative is then to build a stabilization fund, providing governments with their own buffer.
The cost of placing revenues in a stabilization fund is that they need to be held in liquid assets, likely to have a relatively low return. The benefit depends on whether there are alternative ways of handling revenue uncertainty (as outlined above), and on the costs associated with not stabilizing. Are resource revenues a sufficiently large share of economic activity, exports, or government revenue to significantly destabilize the economy? Are other policy tools (such as monetary policy) available and effective to counter economic shocks? Finally, there are difficult issues surrounding the operation of such a fund; above all, how to make the judgement as to when to expand and when to draw down the fund, i.e., on whether prices are abnormally high or low. Typically the decision is based on some moving average of past prices, although one of the most successful stabilization funds, Chile’s Social and Economic Stabilization Fund, uses an independent panel of experts to provide an informed judgement.
Conclusions
Implementation of resource revenue management is context specific and depends on politics as well as economics—but clarity on the economic principles matters. For developing countries, we suggest the key principles are: First, to use a high (and rising) share of resource revenues for building assets, rather than for current consumption. Second, to integrate these with national development plans for building human and physical capital in the country. Efficient domestic investment strategies involve planning ahead, anticipating bottlenecks that will be encountered during a resource boom, and making public investments that will support private sector activity in a resource abundant economy. Third, natural resource funds should be used in support of this domestic investment strategy, rather than as ends in themselves; long-run asset accumulation is better done in the domestic economy than through “inter-generational” offshore funds; parking and stabilization funds are appropriate where they meet well-defined objectives that support domestic economic growth. With the commodity super-cycle coming to a close, now is an appropriate time to prepare for the next cycle.
References
Collier, P. 2010. The Plundered Planet, Oxford University Press, New York.
van der Ploeg F. and A.J. Venables. 2011. “Harnessing Windfall Revenues: Optimal Policies for Resource Rich Developing Economies.” Economic Journal 121:1–31
Venables, A.J. and S. Wills. 2015. “Resource Funds: Stabilizing, Parking and Inter-Generational Transfer” Oxcarre Working Paper. University of Oxford, Oxford, U.K.
Anthony J. Venables and Samuel Wills are both at the Oxford Centre for the Analysis of Resource Rich Economics, Department of Economics, University of Oxford.
