Many countries experience substantial revenue windfalls from natural resources. The consensus is that these should not be consumed immediately but put in a fund, typically a sovereign wealth fund, in order to smooth the benefits across generations and deal with the otherwise adverse effects of Dutch disease and the resource curse. But should they? And if so, why? In a recent column, Cherif and Hasanov (2012) addressed the oil exporters’ dilemma, i.e. how much of the windfall to save and how much to invest. Whereas these authors focus on the sample average for a group of oil-exporting countries, we look at how policy recommendations may differ between particular oil-rich countries and thus address the vexed questions of how to put oil windfalls to good use and how to cope with the historically high volatility of oil prices.
Oil windfalls are, typically, not only anticipated and temporary, but also subject to the considerable volatility of world oil prices. Oil exporters should thus accumulate not only an intergenerational fund to smooth the benefits across different generations, but also a liquidity fund to collect precautionary saving buffers in order to cope with oil price volatility. In addition, a domestic investment fund may be needed to alleviate the burden of capital scarcity. This column argues that there are different rationales for each of these three types of funds, an intergenerational, a liquidity fund and an investment fund, and suggests that three entirely different countries, Norway, Iraq and Ghana should have funds of different types for entirely different reasons. In a recent working paper (van den Bremer & van der Ploeg 2012), we have introduced a simple welfare-based, infinite-horizon model of an oil-exporting economy in order to estimate the magnitudes of the different funds and their determinants. By providing ball-park estimates of the optimal size of these funds as a percentage of non-resource GDP and in dollars per capita for three countries, this column thus tells three different stories.
Do we need an investment fund?
A lesson that is sometimes forgotten is that countries that are well integrated into world capital markets should not spend any part of their oil windfalls on domestic investment projects. If a country can borrow at the risk-free world interest rate, then the only two types of funds that are needed are an intergenerational fund accompanied by a liquidity fund. The separation theorem states that it is optimal to spend not even a single dollar of the oil windfall on domestic investment. This applies to developed countries such as Norway and the Netherlands for which the windfall never provides a rationale for additional investment in projects for which costs outweighed benefits without the windfall. For example, there is no rationale for the large amounts of money from its natural gas windfall that the Netherlands has channelled into public infrastructure projects, such as freight railway lines.
Resource-rich countries facing capital scarcity, on the other hand, may find it attractive to channel part of their windfall towards domestic investment in infrastructure, health and education projects. In this case, the fact that the country pays a premium on its debt leads to breakdown of the separation theorem. The availability of a stream of additional income from the windfall allows for much-needed investment in public capital. This is particularly true for many developing countries. To the extent that such countries also suffer from absorption constraints including planning and implementation lags, there is a rationale for temporarily putting funds in a parking fund until absorption constraints are alleviated.
What determines the size of the intergenerational and the liquidity fund?
The size of the intergenerational fund is larger if the windfall is more temporary and future generations are not expected to be much richer than current generations, since then the need to smooth the windfall across present and future generations is less. Focussing on the effect of the windfall, we have abstracted from borrowing against future growth by smoothing consumption per capita in efficiency units. The size of the liquidity fund, on the other hand, is larger if oil price volatility is bigger, the policymaker is more prudent, the size of the windfall relative to GDP is larger, and the windfall is more permanent. Expected productivity growth curbs the need for precautionary saving, because future oil shocks are easier to absorb if the non-resource part of the economy has grown.
Estimates of intergenerational and liquidity funds
Table 1 below give the optimal sizes of the intergenerational and liquidity funds for Norway, Iraq and Ghana. We have used a power utility function with a coefficient of relative prudence of 3, which implies a coefficient of relative risk aversion of 2. Our estimated stochastic process for world crude oil (gas) prices with mean reversion over the sample period 1960-2011 has a mean price level of $110 per barrel of oil equivalent ($32 for gas), a drift parameter of 0.06 (0.06), and a volatility of 0.26 (0.20). For our base case results we have used a deterministic population growth rate of 0.5%, 2.3% and 2.3%; a deterministic growth rate in non-resource GDP per capita (g) of 1%, 1.8% and 2.5%; and extraction costs of $15, $10 and $25 per barrel of oil equivalent for Norway, Iraq and Ghana, respectively.
Table 1. Optimal sizes of intergenerational and liquidity funds
At about 28% of non-resource GDP in 2012, Norway’s expected windfall is sizeable but declining, tapering off to 15% of non-oil GDP in 2030. The intergenerational fund has to be large, almost 7 times non-resource GDP, in order to give every citizen a permanent consumption annuity of $8,500 per year in 2012 efficiency units or 13% of non-resource GDP, rising at the rate of productivity growth (i.e., 1% per annum). With significant non-resource income, Norway’s liquidity fund is fairly small potatoes (3% of non-oil GDP) and the extent of precautionary saving is small (0.3% of non-oil GDP in 2012).
In contrast, Iraq’s windfall is expected to last much longer and is almost permanent in nature, lasting at least until 2100. In addition, it is gigantic compared with the non-oil economy. Oil rents are 6.5 times non-oil GDP in 2012. This is why Iraq’s intergenerational fund will grow to such a large size, i.e. 173 times non-oil GDP. It will ensure a permanent annuity for every Iraqi citizen of $1,500 in 2012 (and growing at 1.8% per annum). More notably, Iraq’s liquidity fund is 12 times non-oil GDP, much larger in relative terms than that of Norway and a factor to take into account. Because the uncertain oil windfall makes up such a significant part of GDP and is expected to do so for many decades, precautionary saving is a force to reckon with for Iraq.
Ghana’s oil windfall, expected to reach peak production in 2014 at less than 10% of non-oil GDP and then tapering off very quickly, is much smaller than that of Norway and Iraq, both in absolute and relative terms, and expected to last much less long. A liquidity fund is therefore not needed. In Van den Bremer & Van der Ploeg (2012) we use public investment measures of inefficiency to calibrate adjustment costs for public investment. The windfall allows external public and publicly guaranteed debt to be paid off more rapidly, so that the social cost of borrowing falls more rapidly and public investment is stimulated. We show that, if Ghana has to tackle problems of capital scarcity and absorption constraints, the percentage of its windfall that it has to invest in domestic investment projects will be roughly 60% in 2012, rising to 90% in 2018 as capacity adjusts and absorption constraints are alleviated, and steadily falling thereafter.
An intergenerational fund is needed to smooth the benefits of the windfall across generations; a volatility fund is necessary to hedge against oil price volatility, especially as futures and derivatives markets are thin and costly to use, and an investment fund helps to allocate funds to domestic investment in a timely and efficient manner in capital-scarce developing economies with absorption constraints. We offer ballpark estimates of these fund sizes and associated saving and investment needs for three very different countries, i.e. Norway, Iraq and Ghana. This operationalises the different rationales for such funds reviewed by Collier et al. (2010) and Van der Ploeg and Venables (2012) to a stochastic setting with volatile windfalls. Of course, policy for oil exporters should also focus on improving productivity in the tradable sector and reducing volatility through diversifying this sector. Indeed such policies are important in any case, but especially for a country that is susceptible to substantial oil price volatility. Uncertainty regarding the size of the reserves can only strengthen these arguments.
Cherif, R and Hasanov, F (2012) "Oil exporters’ dilemma: How much to save and how much to invest", VoxEU.org, 10 November 2012.
Collier P, Spence M, van der Ploeg F and Venables, A J, (2010), "Managing resource revenues in developing economies", IMF Staff Papers, 57, 1, 84-118.
Van der Ploeg, F and A J Venables, (2012), "Natural resource wealth: the challenge of managing a windfall", Annual Review of Economics, 4, 315-337.
Van den Bremer, T S and Van der Ploeg, F, (2012), "Managing and harnessing volatile oil windfalls", CEPR Discussion Paper 9209.