Annex I Fiscal Policy Sustainability in Nonrenewable Resource-Producing Countries
Among the issues that nonrenewable resource-producing countries need to address are how much of the resource to extract and how much to consume. These two issues can be analyzed separately.
How Much to Extract
Determining the optimal path of extraction of a depletable resource is fraught with uncertainties relating to factors such as future resource prices, costs of extraction, and interest rates; estimates of the resource reserves; and the risk that the resource may become obsolete before the reserves are exhausted. How much of the resource to extract is an issue of portfolio selection under uncertainty, with the added complexity that in the case of some resources such as oil, production decisions are often difficult to reverse, and investment decisions have, in some cases, effects that show up several years later.
An optimal extraction rate will depend in part on assumptions about future resource prices, costs, and interest rates. According to the so-called Hotelling principle, the real price of a nonrenewable natural resource would tend to rise at a rate equal to the real interest rate, in which case, with constant production costs, resource wealth would not be affected by the rate of extraction. In the case of oil, the evidence of the past 50 years (which includes the large positive shocks of the 1970s and 1980s) suggests that the average rate of increase of real oil prices has been less than the real interest rate on U.S. bonds. To the extent that the expected increase in the price of the resource (net of costs) in real terms might be lower than the projected real interest rate, other things being equal it would be advantageous to accelerate extraction.
Decisions about how much to extract, however, would also need to take into account market effects. In particular, in the case of important producers, an analysis of the optimal path of extraction should also include an assessment of the likely effect of production strategies on the output decisions of other producers. Other factors that need to be taken into account include the impact of production levels on extraction costs.
How Much to Consume
A main objective of economic policy in a nonrenewable resource-producing country should be to determine a sustainable level of consumption out of resource wealth. Resource wealth reflects the present discounted value of future government revenues (net of re source-related expenditures) derived from the production of the resource. Estimates of “permanent income” can then be derived based on government net wealth including resource wealth. Governments may “consume” each period up to the real interest they receive on their total wealth. This would set an upper limit on the sustainable nonresource deficit of the government. From the perspective of consumption possibilities, the permanent income model treats resource wealth and financial wealth as equivalents. Hence, the consumption of resource revenues in excess of permanent income would be equivalent to depleting savings held in financial assets.
The permanent income framework can be extended to account for growth of the population. If the government aims at preserving its wealth per capita, it must adjust its consumption for the expanding population. The reason for maintaining per capita wealth is that it allows the same per capita income stream to be kept over time, and thus constant per capita government expenditures.
Since the development of the nonresource sector provides an additional source of income for the government, revenue generated by an expanding nonresource sector could over time replace revenue from exploiting the resource. Thus, to the extent that nonresource GDP per capita is projected to grow, in allocating consumption over time the government could take into account the expected future growth in nonresource revenue. However, additional nonresource revenue must be generated by growth in nonresource GDP—not by heavier tax burdens—and such growth must be assessed cautiously. Moreover, this could prescribe redistributing resources from future to current generations, but borrowing against future income may be undesirable, since future resources are inherently uncertain.
The estimation of resource wealth at any given point in time is subject to considerable uncertainty. Estimates must rely on assumptions about resource reserves, future prices, production costs, extraction rates, and interest rates, as well as the possible obsolescence of the resource at some point in the future. The behavior of prices, as well as new information on other determinants of resource wealth, could lead to significant adjustments in the wealth estimates. The permanent negative oil shock of 1986 is instructive in this respect—its severity led to questions about the solvency of a number of oil-producing countries, making the existing policy stance unsustainable in the absence of corrective action. Therefore, resource wealth estimates underlying long-run policy decisions must be adapted frequently to new information about the factors mentioned above; sensitivity analyses should be performed; and the issue of sustainability must be regularly revisited.
With sizable uncertainty surrounding resource prices and income, reserves, and other factors, there is a need to apply possibly substantial precautionary correction factors to the estimates. The adjusted estimates could be used to help determine a sustainable fiscal policy in terms of prudent limits to the nonresource fiscal deficit.
Finally, the approach of using estimates of permanent income adjusted for risk to help determine a proper stance of fiscal policy in the medium and long term needs to be supplemented by other shorter-term considerations. Adjustment costs are discussed in the main sections of this paper. Other factors that need to be taken into account include the level of public debt and, in the short term, the economic cycle and financing constraints, together with their implications for domestic interest rates and for the adequate access to credit by the private sector. In certain situations these factors could suggest the need for a more restrained nonresource fiscal stance than required by fiscal sustainability adjusted for risk.
Annex II An Empirical Analysis of Nonrenewable Resource Export Earnings and Government Expenditure
This annex describes the empirical exercises presented in Section IV in greater detail. The following sections discuss the data, describe the model selection process, and present the structural break tests.
Data
The main variables used were central government expenditure and non renewable resource export earnings, both expressed in real per capita terms (consumer prices were used as the price deflators). The data, unless otherwise noted, were taken from the IMF’s World Economic Outlook database.1 Annual data were used, and the sample for each country was determined by data availability (1963 was the earliest observation). However, the sample for a given country does not begin until the nonrenewable resource export earnings exceed If) percent of central government expenditure: once there is one observation over this threshold, the sample starts, even if subsequent observations fall below the 10 percent ratio. If available, observations prior to the 10 percent threshold were used when lagged values were required. Finally, the sample for Venezuela ends in 1998 in order to exclude the first year of the oil fund, and that for Papua New Guinea begins in 1974 in order to exclude the two years without an NRF.
The sample consisted of 12 countries, 5 countries with an NRF and 7 other countries selected to serve as a comparator group. Basic summary statistics for the ratio of nonrenewable resource export earnings to central government spending are presented in Table A2.1. To obtain results that are directly comparable across countries, the following regression was run for each country,
ΔGt = b0 + b1ΔXt + b2ΔXt−1.
where Δ is the first difference operator, G is central government expenditure, and X is nonrenewable resource export earnings. The results from this regression were then used to calculate the impact of a 50 percent increase in the nonrenewable resource price on central government expenditure.
Selected Countries: Summary Sample Information
Evaluated at the average nonrenewable resource export earnings and government expenditure.
The test statistic from an augmented Dickey-Fuller test assuming zero (ADF(0)) or first order (ADF(1)) serial correlation. The 10 percent critical values ranged from 2.6 to 2.7 with no trend and 3.2 to 3.3 with a trend. Bold indicates that die hypothesis of a unit root could be rejected at the 10 percent level.
The number of cointegrating equations is reported, AR(0) and AR(1) refer to the level of autocorrelation assumed.
Only the estimates on the nonrewable resource export earnings term are reported, other explanatory variables may have been included.
Selected Countries: Summary Sample Information
| Countries with an NRF | Countries without an NRF | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Chile | Kuwait | Norway | Oman | Papua New Guinea | Algeria | Bahrain | Mexico | Arabia Saudi | United Arab Emirates | United Kingdom | Venezuela | |||
| Basic information | ||||||||||||||
| Sample | 1969–99 | 1965–90 | 1976–99 | 1965–99 | 1974–99 | 1979–99 | 1965–99 | 1980–99 | 1965–99 | 1980–99 | 1983–99 | 1965–98 | ||
| Average X/G (percent) | 31.4 | 141.0 | 30.7 | 73.2 | 71.4 | 68.2 | 79.3 | 16.4 | 83.0 | 101.0 | 5.0 | 86.2 | ||
| High X / G (percent) | 61.2 | 370.5 | 47.5 | 128.9 | 115 | 109.5 | 1895 | 245 | 209.2 | 197.6 | 11.6 | 165.4 | ||
| Low X / G (percent) | 6.2 | 61.9 | 10.3 | 12.5 | 38.5 | 23.6 | 38.0 | 10.3 | 41.2 | 54.1 | 2.3 | 48.5 | ||
| Regression results (dependent variable is ΔGt) | ||||||||||||||
| b0 (constant) | .041 | −.0001 | .015* | .00002 | −.00006 | .001 | .00009 | .00005 | −.0005 | −.019** | .0009* | −.02 | ||
| (.036) | (.0006) | (.005) | (.0002) | (.00005) | (.005) | (.0002) | (000.1) | (−003) | (.009) | (002) | (.04) | |||
| b1 (Coeffficient on ΔXt) | −.042 | −.12** | .04 | .31* | −.04 | .05 | −.09 | .98** | .21** | .10* | .55 | .08 | ||
| (.135) | (0.05) | (.10) | (.08) | (.08) | (.10) | (.11) | (.46) | (.05) | (.03) | (.36) | (.07) | |||
| b2(Coefficient on ΔXt-1) | −.358** | .12** | −.15 | .30* | .13** | .21* | .34* | .24 | .05 | .06** | .61 | .16** | ||
| (.136) | (.05) | (.10) | (.08) | (.06) | (.10) | (.11) | (.45) | (.05) | (.03) | (.36) | (.07) | |||
| Observations | 31 | 26 | 24 | 35 | 26 | 21 | 35 | 20 | 35 | 20 | 17 | 34 | ||
| R-squared | .2 | .38 | .11 | .47 | .16 | .24 | .22 | .23 | .34 | .52 | .26 | .15 | ||
| Impact of SO percent increase in nonrenewable resource price (in percent of average G)1 | ||||||||||||||
| Estimated impact | −6.2 | −0.2 | −1.7 | 24.2 | 3.0 | 5.7 | 9.9 | 10.2 | 11.5 | 9.3 | 2.3 | 10.3 | ||
| Standard deviation | 3.1 | 4.9 | 2.2 | 4.5 | 3.2 | 4.1 | 5.8 | 5.3 | 3.2 | 2.2 | 1.2 | 4.7 | ||
| Tests (no trend in data) | ||||||||||||||
| Unit root tests2 | ||||||||||||||
| X, ADF(0) | −1.69 | −2.02 | −1.98 | −1.97 | −2.05 | −3.23 | −1.61 | −2.26 | −2.00 | −2.73 | −1.39 | −3.00 | ||
| X, ADF(1) | −1.61 | −1.90 | −1.94 | −1.96 | −2.52 | −4.35 | −2.14 | −2.54 | −2.45 | −2.84 | −1.37 | −2.49 | ||
| G, ADF(0) | −0.46 | −2.26 | −0.33 | −1.65 | −1.68 | −2.72 | −1.80 | −1.41 | −1.17 | −1.28 | −0.97 | −1.39 | ||
| G, ADF(1) | −0.75 | −2.17 | −0.32 | −2.14 | −1.96 | −2.95 | −2.02 | −1.84 | −1.37 | −1.24 | −1.24 | −0.86 | ||
| Cointegration tests3 | ||||||||||||||
| AR(0) | None | None | None | None | None | 2 | 1 | None | None | 1 | None | None | ||
| AR(1) | None | None | None | None | 1 | 2 | 1 | None | None | 2 | None | None | ||
| Tests (trend in data) | ||||||||||||||
| Unit root tests2 | ||||||||||||||
| X, ADF(0) | −2.91 | −2.18 | −2.35 | −1.57 | −3.05 | −3.03 | −1.58 | −3.60 | −2.03 | −5.59 | −1.69 | −3.05 | ||
| X, ADF(1) | −3.14 | −2.06 | −2.44 | −1.44 | −4.57 | −4.25 | −2.08 | −3.74 | −2.48 | −6.64 | −1.76 | −2.55 | ||
| G, ADF(0) | −0.57 | −2.16 | −2.21 | −1.01 | −3.04 | −2.53 | −1.64 | −3.15 | −1.20 | −1.45 | −0.89 | −2.85 | ||
| G, ADF(1) | −0.59 | −2.11 | −2.52 | −1.80 | −3.87 | −2.74 | −1.89 | −3.40 | −1.39 | −1.60 | −2.94 | −2.36 | ||
| Cointegration tests 3 | ||||||||||||||
| AR(0) | None | 1 | None | 2 | None | 2 | 2 | 1 | None | 1 | None | 1 | ||
| AR(1) | None | None | None | None | 1 | 2 | 1 | 1 | 1 | 2 | None | None | ||
| Preferred specification4 | ||||||||||||||
| L=level; D=diffs.;T =trend | 0 | D | D-T | D | D | L | D | D | O | L | D-T | D | ||
| −.12** | .31* | .99** | .21* | .34* | ||||||||||
| a2(Xt) or b1 (ΔXt) | n.a. | −(05) | n.a. | (.08) | n.a. | n.a. | n.a. | (.43) | (.05) | (.04) | n.a. | n.a. | ||
| a3(Xt-1) or b2 (ΔXt-1) | −.33** | .12** | −.15 | .30* | .12*** | .21* | .32* | .07** | −.59 | .14** | ||||
| (.13) | (.05) | (.10) | (.08) | (.06) | (0.7) | (.11) | n.a. | n.a. | (0.3) | (.38) | (.07) | |||
| Observations | 31 | 26 | 24 | 35 | 26 | 21 | 35 | 20 | 36 | 20 | 17 | 34 | ||
| R-squared | .16 | .33 | .1 | .47 | .11 | .58 | .2 | .21 | .32 | .97 | .14 | .1 | ||
Evaluated at the average nonrenewable resource export earnings and government expenditure.
The test statistic from an augmented Dickey-Fuller test assuming zero (ADF(0)) or first order (ADF(1)) serial correlation. The 10 percent critical values ranged from 2.6 to 2.7 with no trend and 3.2 to 3.3 with a trend. Bold indicates that die hypothesis of a unit root could be rejected at the 10 percent level.
The number of cointegrating equations is reported, AR(0) and AR(1) refer to the level of autocorrelation assumed.
Only the estimates on the nonrewable resource export earnings term are reported, other explanatory variables may have been included.
Selected Countries: Summary Sample Information
| Countries with an NRF | Countries without an NRF | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Chile | Kuwait | Norway | Oman | Papua New Guinea | Algeria | Bahrain | Mexico | Arabia Saudi | United Arab Emirates | United Kingdom | Venezuela | |||
| Basic information | ||||||||||||||
| Sample | 1969–99 | 1965–90 | 1976–99 | 1965–99 | 1974–99 | 1979–99 | 1965–99 | 1980–99 | 1965–99 | 1980–99 | 1983–99 | 1965–98 | ||
| Average X/G (percent) | 31.4 | 141.0 | 30.7 | 73.2 | 71.4 | 68.2 | 79.3 | 16.4 | 83.0 | 101.0 | 5.0 | 86.2 | ||
| High X / G (percent) | 61.2 | 370.5 | 47.5 | 128.9 | 115 | 109.5 | 1895 | 245 | 209.2 | 197.6 | 11.6 | 165.4 | ||
| Low X / G (percent) | 6.2 | 61.9 | 10.3 | 12.5 | 38.5 | 23.6 | 38.0 | 10.3 | 41.2 | 54.1 | 2.3 | 48.5 | ||
| Regression results (dependent variable is ΔGt) | ||||||||||||||
| b0 (constant) | .041 | −.0001 | .015* | .00002 | −.00006 | .001 | .00009 | .00005 | −.0005 | −.019** | .0009* | −.02 | ||
| (.036) | (.0006) | (.005) | (.0002) | (.00005) | (.005) | (.0002) | (000.1) | (−003) | (.009) | (002) | (.04) | |||
| b1 (Coeffficient on ΔXt) | −.042 | −.12** | .04 | .31* | −.04 | .05 | −.09 | .98** | .21** | .10* | .55 | .08 | ||
| (.135) | (0.05) | (.10) | (.08) | (.08) | (.10) | (.11) | (.46) | (.05) | (.03) | (.36) | (.07) | |||
| b2(Coefficient on ΔXt-1) | −.358** | .12** | −.15 | .30* | .13** | .21* | .34* | .24 | .05 | .06** | .61 | .16** | ||
| (.136) | (.05) | (.10) | (.08) | (.06) | (.10) | (.11) | (.45) | (.05) | (.03) | (.36) | (.07) | |||
| Observations | 31 | 26 | 24 | 35 | 26 | 21 | 35 | 20 | 35 | 20 | 17 | 34 | ||
| R-squared | .2 | .38 | .11 | .47 | .16 | .24 | .22 | .23 | .34 | .52 | .26 | .15 | ||
| Impact of SO percent increase in nonrenewable resource price (in percent of average G)1 | ||||||||||||||
| Estimated impact | −6.2 | −0.2 | −1.7 | 24.2 | 3.0 | 5.7 | 9.9 | 10.2 | 11.5 | 9.3 | 2.3 | 10.3 | ||
| Standard deviation | 3.1 | 4.9 | 2.2 | 4.5 | 3.2 | 4.1 | 5.8 | 5.3 | 3.2 | 2.2 | 1.2 | 4.7 | ||
| Tests (no trend in data) | ||||||||||||||
| Unit root tests2 | ||||||||||||||
| X, ADF(0) | −1.69 | −2.02 | −1.98 | −1.97 | −2.05 | −3.23 | −1.61 | −2.26 | −2.00 | −2.73 | −1.39 | −3.00 | ||
| X, ADF(1) | −1.61 | −1.90 | −1.94 | −1.96 | −2.52 | −4.35 | −2.14 | −2.54 | −2.45 | −2.84 | −1.37 | −2.49 | ||
| G, ADF(0) | −0.46 | −2.26 | −0.33 | −1.65 | −1.68 | −2.72 | −1.80 | −1.41 | −1.17 | −1.28 | −0.97 | −1.39 | ||
| G, ADF(1) | −0.75 | −2.17 | −0.32 | −2.14 | −1.96 | −2.95 | −2.02 | −1.84 | −1.37 | −1.24 | −1.24 | −0.86 | ||
| Cointegration tests3 | ||||||||||||||
| AR(0) | None | None | None | None | None | 2 | 1 | None | None | 1 | None | None | ||
| AR(1) | None | None | None | None | 1 | 2 | 1 | None | None | 2 | None | None | ||
| Tests (trend in data) | ||||||||||||||
| Unit root tests2 | ||||||||||||||
| X, ADF(0) | −2.91 | −2.18 | −2.35 | −1.57 | −3.05 | −3.03 | −1.58 | −3.60 | −2.03 | −5.59 | −1.69 | −3.05 | ||
| X, ADF(1) | −3.14 | −2.06 | −2.44 | −1.44 | −4.57 | −4.25 | −2.08 | −3.74 | −2.48 | −6.64 | −1.76 | −2.55 | ||
| G, ADF(0) | −0.57 | −2.16 | −2.21 | −1.01 | −3.04 | −2.53 | −1.64 | −3.15 | −1.20 | −1.45 | −0.89 | −2.85 | ||
| G, ADF(1) | −0.59 | −2.11 | −2.52 | −1.80 | −3.87 | −2.74 | −1.89 | −3.40 | −1.39 | −1.60 | −2.94 | −2.36 | ||
| Cointegration tests 3 | ||||||||||||||
| AR(0) | None | 1 | None | 2 | None | 2 | 2 | 1 | None | 1 | None | 1 | ||
| AR(1) | None | None | None | None | 1 | 2 | 1 | 1 | 1 | 2 | None | None | ||
| Preferred specification4 | ||||||||||||||
| L=level; D=diffs.;T =trend | 0 | D | D-T | D | D | L | D | D | O | L | D-T | D | ||
| −.12** | .31* | .99** | .21* | .34* | ||||||||||
| a2(Xt) or b1 (ΔXt) | n.a. | −(05) | n.a. | (.08) | n.a. | n.a. | n.a. | (.43) | (.05) | (.04) | n.a. | n.a. | ||
| a3(Xt-1) or b2 (ΔXt-1) | −.33** | .12** | −.15 | .30* | .12*** | .21* | .32* | .07** | −.59 | .14** | ||||
| (.13) | (.05) | (.10) | (.08) | (.06) | (0.7) | (.11) | n.a. | n.a. | (0.3) | (.38) | (.07) | |||
| Observations | 31 | 26 | 24 | 35 | 26 | 21 | 35 | 20 | 36 | 20 | 17 | 34 | ||
| R-squared | .16 | .33 | .1 | .47 | .11 | .58 | .2 | .21 | .32 | .97 | .14 | .1 | ||
Evaluated at the average nonrenewable resource export earnings and government expenditure.
The test statistic from an augmented Dickey-Fuller test assuming zero (ADF(0)) or first order (ADF(1)) serial correlation. The 10 percent critical values ranged from 2.6 to 2.7 with no trend and 3.2 to 3.3 with a trend. Bold indicates that die hypothesis of a unit root could be rejected at the 10 percent level.
The number of cointegrating equations is reported, AR(0) and AR(1) refer to the level of autocorrelation assumed.
Only the estimates on the nonrewable resource export earnings term are reported, other explanatory variables may have been included.
Model Selection
To further assess the statistical significance of nonrenewable resource export earnings in explaining central government expenditure, a country-specific specification was selected.
While there are merits in using the same equation for each country, as above, alternative specifications could perform better for a given country. The issues considered below are whether to (1) use first differences or levels; (2) allow for a time trend (drift); and (3) include lagged, contemporaneous, or both nonrenewable resource export earning terms. The decision whether to use levels or first differences depends on the assumptions about unit roots. Given the size of the samples, it is difficult to make conclusive statements about the presence of unit roots (nonstationarity) in the data. Nonetheless, there is reason to suspect, based on the properties of oil prices, that both central government expenditure and nonrenewable resource export earnings could have a unit root. Indeed, for most countries the hypothesis of a unit root could not be rejected for either nonrenewable resource export earnings or central government expenditure (see Table A2.1), suggesting that the regression should usually be run in first differences.2 A caveat for using the first-difference specification, however, is that the nonrenewable resource export earnings and government expenditure should not be cointegrated. In most cases, the hypothesis of cointegration could be rejected.3
Corresponding to the results from the unit root tests, regressions were run using either levels or first differences. Because the unit root tests assumed no trend in the data, the regressions also assume no trend.4 To start with, both the lagged and contemporaneous nonrenewable resource export earnings are included. If both are significant (at the 10 percent level), then this is the preferred specification. Otherwise, the less significant of the two variables is excluded, with the preferred equation then including either just lagged or just contemporaneous nonrenewable resource export earnings. The contemporaneous term would be significant if the government is able to immediately increase spending. The lagged term could be significant if strict adherence to a budget constrained spending until the next budget cycle or if the exports occurred toward the end of the year causing the spending to spill into the following year. Once the preferred specification was chosen, the possibility of a trend was examined by repeating the preferred specification with either a constant (first-differences) or time trend (levels) included. If the trend term was significant, then the entire selection process, including the unit root and cointegration tests, was repeated with a trend included. Finally, for all but two countries, the preferred specification suggested that nonrenewable resource export earnings are a statistically significant determinant of government expenditure (see Table A2.1).
Structural Break Tests
For the countries with an NRF, a key question is whether the relationship between government expenditure and nonrenewable resource export earnings changed with the establishment of the NRF.5 Three tests are performed comparing this relationship in the pre- and post-NRF periods: (1) a Chow test for structural stability; (2) a test of the statistical significance of a post-NRF dummy variable; and (3) a test of the statistical significance of an interaction variable (the post-NRF dummy variable multiplied by the nonrenewable resource export earnings variable(s)). While the tests are in many ways complementary, the Chow test looks for a general regime change whereas the other two tests examine a specific type of regime change. The tests are performed on the preferred specification (as selected by the procedure described above), but are also repeated on alternative specifications to provide a check of robustness.
For each of the countries, the results suggest that the establishment of the NRF did not have a significant impact on government spending (Table A2.2). Some key features of the results are noted below:
-
Chile. The results actually suggest that spending may have increased following the establishment of the NRF (column 2), a result that is consistent with the visual inspection of government expenditure in Chile. The 1990s, however, were also a time of strong real growth, which is the likely cause for the growth in government spending. Including the (lagged) change in real per capita GDP confirms this, since the dummy is no longer significant (column 5).
-
Kuwait. The results for Kuwait are difficult to interpret, since the preferred specification indicates that a change in oil export earnings coincides with an immediate decrease in spending that is completely reversed in the next year—for a net effect of zero. Nonetheless, there is no evidence of structural break (columns 2 and 3). The alternative specification, which uses just lagged oil export earnings, yields the same qualitative result of no structural break. Using just contemporaneous oil export earnings (results not reported) yields the same result.
-
Norway. The preferred specification indicates that there was no change following the establishment of the fund, but only the constant is ever significant. The alternative specification (using levels) is more interesting in that the oil export term is always significant, but there is still, however, no evidence of a structural break.
-
Oman. For Oman, both lagged and contemporaneous oil exports are positive and significant, suggesting that the oil export earnings are spent over two periods. There is no evidence that the establishment of a fund coincided with a change in expenditure in the preferred or alternative specification. As with Kuwait, the same results were obtained using just contemporaneous oil export earnings (results not reported).
Selected Countries: Impact of Nonrenewable Resource Funds
(Dependent variable is central government expenditure)
Selected Countries: Impact of Nonrenewable Resource Funds
(Dependent variable is central government expenditure)
| Preferred Specification | Alternative Specification | |||||||
|---|---|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | |||
| Chile (fund starts in 1986; sample 1969–99; 31 observations) | ||||||||
| ΔXt-1 | −.33** | −.36* | −.45** | ΔXt-1 | −.35* | −.36* | −.40** | |
| (0.13) | (.13) | (.20) | (.13) | (.13) | (.19) | |||
| Dummy | n.a. | .11** | n.a. | Dummy | n.a. | .06 | n.a. | |
| (.05) | (.07) | |||||||
| Dummy*ΔXt-1 | n.a. | n.a. | .21 | Dummy*ΔXt-1 | n.a. | n.a. | .13 | |
| (.27) | (.26) | |||||||
| ΔYt-1 | n.a. | n.a. | n.a. | ΔYt-1 | .003** | .002 | .003** | |
| (.002) | (.002) | (.002) | ||||||
| Chow test | .44 | n.a. | n.a. | Chow test | .60 | n.a. | n.a. | |
| R-squared | .16 | 0.28 | 0.18 | R-squared | .28 | .30 | .29 | |
| Kuwait (fund starts in 1976; sample 1965–90; 26 observations) | ||||||||
| ΔXt | −.12** | −.12** | −.07 | ΔXt | n.a. | n.a. | n.a. | |
| (.05) | (.05) | (.06) | ||||||
| ΔXt-1 | .12** | .12** | .14*** | ΔXt-1 | .13** | .13** | .16** | |
| (.05) | (.05) | (.07) | (.05) | (.05) | (.07) | |||
| Dummy | n.a. | .0001 | n.a. | Dummy | n.a. | .0005 | n.a. | |
| (.0008) | (.0008) | |||||||
| Dummy*ΔXt | n.a. | n.a. | −.12 | Dummy*ΔXt | n.a. | n.a. | n.a. | |
| (.10) | ||||||||
| Dummy*ΔXt-1 | n.a. | n.a. | −.02 | Dummy*ΔXt-1 | n.a. | n.a. | −.06 | |
| (.09) | (.10) | |||||||
| Chow test | .45 | n.a. | n.a. | Chow test | 0.58 | n.a. | n.a. | |
| R-squared | .38 | 0.38 | 0.42 | R-squared | 0.2 | 0.21 | 0.21 | |
| Norway (fund starts in 1991; sample 1976–99; 24 observations) | ||||||||
| Constant | .016* | .017* | .016* | Trend | .009* | .009* | .009* | |
| (.005) | (.006) | (.005) | (.003) | (.003) | (.003 | |||
| ΔXt-1 | −.15 | −.16 | −.17 | ΔXt-1 | −0.18** | −0.18** | −0.18** | |
| (.10) | (.10) | (.13) | (.07) | (.07) | (.07) | |||
| Dummy | n.a. | −.004 | n.a. | Dummy | n.a. | .01 | n.a. | |
| (.010) | (.02) | |||||||
| Dummy*ΔXt-1 | n.a. | n.a. | .06 | Dummy*ΔXt-1 | n.a. | n.a. | .02 | |
| (.21) | (.07) | |||||||
| Constant | .12** | .14** | .13** | |||||
| (.04) | (.07) | (.06) | ||||||
| G t-1 | 0.51* | 0.47** | 0.49** | |||||
| (0.16) | (0.19) | (0.18) | ||||||
| Chow test | .90 | n.a. | n.a. | Chow test | 0.8 | n.a. | n.a. | |
| R-squared | .10 | .10 | .10 | R-squared | .97 | .97 | .97 | |
| Oman (fund starts in 1980; sample 1964–99; 36 observations) | ||||||||
| ΔXt | .31* | .31* | .23*** | ΔXt | n.a. | n.a. | n.a. | |
| (.08) | (.08) | (.13) | ||||||
| ΔXt-1 | .30* | .30* | .27*** | ΔXt-1 | .27* | .28* | .30*** | |
| (.08) | (.08) | (.13) | (.09) | (.09) | (.13) | |||
| Dummy | n.a. | .00004 | n.a. | Dummy | n.a. | .00003 | n.a. | |
| .00003 | (.00003) | |||||||
| Dummy*ΔXt | n.a. | n.a. | .15 | Dummy*ΔXt | n.a. | n.a. | n.a. | |
| (.16) | ||||||||
| Dummy*ΔXt-1 | n.a. | n.a. | .07 | Dummy*ΔXt-1 | n.a. | n.a. | −.04 | |
| (.16) | (.20) | |||||||
| Chow test | .60 | n.a | n.a. | Chow test | .84 | n.a. | n.a. | |
| R-squared | .47 | 0.47 | 0.49 | R-squared | .20 | 0.21 | 0.21 | |
Selected Countries: Impact of Nonrenewable Resource Funds
(Dependent variable is central government expenditure)
| Preferred Specification | Alternative Specification | |||||||
|---|---|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | |||
| Chile (fund starts in 1986; sample 1969–99; 31 observations) | ||||||||
| ΔXt-1 | −.33** | −.36* | −.45** | ΔXt-1 | −.35* | −.36* | −.40** | |
| (0.13) | (.13) | (.20) | (.13) | (.13) | (.19) | |||
| Dummy | n.a. | .11** | n.a. | Dummy | n.a. | .06 | n.a. | |
| (.05) | (.07) | |||||||
| Dummy*ΔXt-1 | n.a. | n.a. | .21 | Dummy*ΔXt-1 | n.a. | n.a. | .13 | |
| (.27) | (.26) | |||||||
| ΔYt-1 | n.a. | n.a. | n.a. | ΔYt-1 | .003** | .002 | .003** | |
| (.002) | (.002) | (.002) | ||||||
| Chow test | .44 | n.a. | n.a. | Chow test | .60 | n.a. | n.a. | |
| R-squared | .16 | 0.28 | 0.18 | R-squared | .28 | .30 | .29 | |
| Kuwait (fund starts in 1976; sample 1965–90; 26 observations) | ||||||||
| ΔXt | −.12** | −.12** | −.07 | ΔXt | n.a. | n.a. | n.a. | |
| (.05) | (.05) | (.06) | ||||||
| ΔXt-1 | .12** | .12** | .14*** | ΔXt-1 | .13** | .13** | .16** | |
| (.05) | (.05) | (.07) | (.05) | (.05) | (.07) | |||
| Dummy | n.a. | .0001 | n.a. | Dummy | n.a. | .0005 | n.a. | |
| (.0008) | (.0008) | |||||||
| Dummy*ΔXt | n.a. | n.a. | −.12 | Dummy*ΔXt | n.a. | n.a. | n.a. | |
| (.10) | ||||||||
| Dummy*ΔXt-1 | n.a. | n.a. | −.02 | Dummy*ΔXt-1 | n.a. | n.a. | −.06 | |
| (.09) | (.10) | |||||||
| Chow test | .45 | n.a. | n.a. | Chow test | 0.58 | n.a. | n.a. | |
| R-squared | .38 | 0.38 | 0.42 | R-squared | 0.2 | 0.21 | 0.21 | |
| Norway (fund starts in 1991; sample 1976–99; 24 observations) | ||||||||
| Constant | .016* | .017* | .016* | Trend | .009* | .009* | .009* | |
| (.005) | (.006) | (.005) | (.003) | (.003) | (.003 | |||
| ΔXt-1 | −.15 | −.16 | −.17 | ΔXt-1 | −0.18** | −0.18** | −0.18** | |
| (.10) | (.10) | (.13) | (.07) | (.07) | (.07) | |||
| Dummy | n.a. | −.004 | n.a. | Dummy | n.a. | .01 | n.a. | |
| (.010) | (.02) | |||||||
| Dummy*ΔXt-1 | n.a. | n.a. | .06 | Dummy*ΔXt-1 | n.a. | n.a. | .02 | |
| (.21) | (.07) | |||||||
| Constant | .12** | .14** | .13** | |||||
| (.04) | (.07) | (.06) | ||||||
| G t-1 | 0.51* | 0.47** | 0.49** | |||||
| (0.16) | (0.19) | (0.18) | ||||||
| Chow test | .90 | n.a. | n.a. | Chow test | 0.8 | n.a. | n.a. | |
| R-squared | .10 | .10 | .10 | R-squared | .97 | .97 | .97 | |
| Oman (fund starts in 1980; sample 1964–99; 36 observations) | ||||||||
| ΔXt | .31* | .31* | .23*** | ΔXt | n.a. | n.a. | n.a. | |
| (.08) | (.08) | (.13) | ||||||
| ΔXt-1 | .30* | .30* | .27*** | ΔXt-1 | .27* | .28* | .30*** | |
| (.08) | (.08) | (.13) | (.09) | (.09) | (.13) | |||
| Dummy | n.a. | .00004 | n.a. | Dummy | n.a. | .00003 | n.a. | |
| .00003 | (.00003) | |||||||
| Dummy*ΔXt | n.a. | n.a. | .15 | Dummy*ΔXt | n.a. | n.a. | n.a. | |
| (.16) | ||||||||
| Dummy*ΔXt-1 | n.a. | n.a. | .07 | Dummy*ΔXt-1 | n.a. | n.a. | −.04 | |
| (.16) | (.20) | |||||||
| Chow test | .60 | n.a | n.a. | Chow test | .84 | n.a. | n.a. | |
| R-squared | .47 | 0.47 | 0.49 | R-squared | .20 | 0.21 | 0.21 | |
For Chile, the copper export earnings series is from Spilimbergo(1999), with the underlying source being the Central Bank of Chile. For Papua New Guinea, the source for the mineral export series is the Bank of Papua New Guinea, Quarterly Economic Review.
For the United Arab Emirates and Venezuela there is some evidence of unit root in government expenditure but not in nonrenewable resource export earnings. A subjective assessment in these two cases suggests using levels for the United Arab Emirates and different for Venezuela…
The exception is Bahrain, but the cointegrating relationship suggests a similar impact of oil export earnings on government expenditure. The presence of two cointegrating relationships for Algeria and United Arab Emirates is consistent with the series being stationary.
Specifically, Gt = a0 + a1Gt-1 + a2Xt + a3Xt-1 was used for levels and ΔGt = b2ΔXt + b3ΔXt − 1 for first differences.
Papua New Guinea was excluded owing to insufficient pre-NRF data.
Annex III Commodity Risk Markets
A possible alternative and complementary approach to stabilization funds is for governments to reduce the uncertainty and volatility of nonrenewable resource revenues directly through commodity risk markets. This appendix discusses, from this viewpoint, hedging experience and its possible uses and constraints. Commodity hedging instruments and strategies are summarized in Box A3.1. This annex focuses on oil, given its importance as a non renewable resource, but the discussion is relevant for other nonrenewable commodities.
Possible Uses of Hedging
Hedging can reduce the uncertainty and volatility of a commodity revenue stream.1 The possible use and benefit of hedging depends on the extent to which the uncertainty and volatility are reduced by the hedging strategy and how far they had created fiscal and macroeconomic problems (discussed in Section II).2 More specifically, government commodity hedging operations could have the following uses.
Reducing the volatility and uncertainty of the revenue stream could allow budgeting to become more realistic and certain. For example, if when preparing the budget for 2001, the state-owned oil exporter had already sold its 2001 output for $26 per barrel via the futures market, the budget could be prepared on the basis of that actual price. This would also preclude any unexpected over- or undershooting of oil revenue during 2001, which in turn could help to prevent sharp expenditure changes.
Hedging could provide “disaster” insurance. Governments may be able to manage small fluctuations in the price received for their resource output compared with the price budgeted. But a massive shortfall is difficult to deal with. The purchase of options could offer insurance against such a disaster scenario.3 Purchasing options is costly, but insofar as the price at which the government wishes to guarantee its output is far below the market-expected price, the cost of buying the option would likely be low in relation to the potential gains.
Hedging could foster smoother fiscal adjustment to permanent price shocks. If a government were to sell its 2001 output in 2000, a sharp fall in the spot price in 2001 would not affect government revenue in that year. To the extent that the fall in the spot price was reflected in lower futures prices too, then as the government started selling its 2002 output during 2001, the government could see that its revenue in 2002 would be lower and set its budget on that basis. Should the fall in the spot price not be reflected in futures prices (for example, if the market judged the shock to be temporary), there would be no revenue impact in either 2001 or 2002.
A number of other possible uses of hedging can also be envisaged. Because the government could become less vulnerable to shocks, it could keep a lower level of liquid foreign exchange reserves. It could also reduce the need to borrow heavily when the commodity price falls. Assuming that the hedging is in foreign exchange, balance of payments flows could similarly become more certain and less volatile (for instance, a sharp increase in the spot price of oil would not lead to a sudden surge in foreign exchange inflows). With a more predictable revenue stream, government creditworthiness could improve, and thus lead to lower-cost borrowing. The political pressure to spend when the spot oil price increases sharply may be lessened, since this may not immediately translate into higher revenue. With the budget more realistic and based on a factual, rather than a predicted price, transparency could also be enhanced, and forecasting simplified.
Hedging Instruments: A Summary
Commodity producers bear the risk that the spot price of the commodity they produce may be different in the future. Commodity risk markets could allow producers to reduce this price risk via exchange-traded instruments or over-the-counter instruments.
Exchange-Traded Instruments
A futures contract is a legally binding obligation for the holder of the contract to buy or sell a particular commodity at a predetermined price and location at a specific date in the future. The location, commodity types, and quantity are standardized. Futures contracts rarely give rise to physical delivery.
The purchase of an options contract gives the purchaser the right, but not the obligation, to buy or sell a futures contract at a given price. A buyer of an options contract pays a premium to the seller at the time of the purchase. The buyer has a potential maximum loss equal to the premium, but potentially unlimited gains. An options contract to buy a futures contract is known as a “call” option; an options contract to sell a futures contract is known as a “put” option.
Over-the-Counter Instruments
Forward contracts are agreements to sell or buy a certain product at a certain future time at a preset price. They generally give rise to physical deliveries.
Commodity swaps are basically agreements between two parties to buy or sell a commodity at a fixed price for many periods in the future. Basic (“plain vanilla”) commodity swaps involve one party exchanging a fixed price for a floating price. Whereas a futures contract or a forward relates to the price of one transaction in the future, a swap relates to many transactions, often for much longer into the future.
Commodity bonds or loans are bonds or loans with payments (of principal or interest or both) linked to commodity prices.
Hybrids are combinations of other instruments; for example, a “swaption” is an option to buy or sell a swap.
The potential uses of hedging do not include that of fostering government savings. Like funds, hedging does not directly affect expenditure and thus does not affect saving for future generations, and it is no substitute for sound fiscal policy. Moreover, since markets for long-term hedging instruments are limited (see below), hedging cannot reduce the uncertainty associated with the largest share of nonrenewable resource wealth, namely revenue streams beyond the periods immediately ahead. Also, hedging may lead to problems and constraints of its own.
To help developing countries deal with commodity price volatility and uncertainty, the World Bank recently convened an International Task Force (ITF) comprising international institutions, producers’ and consumers’ organizations, major commodity exchanges, and commodity trading firms (World Bank, 1999). Among other proposals, the ITF recommended the creation of an international intermediary to facilitate access to market-based commodity price insurance.
Country Experience
Information on actual use of commodity risk markets by nonrenewable commodity-exporting governments (or their state-owned exporting enterprises) is limited. This partly reflects client confidentiality and an unwillingness of producers to reveal market-sensitive information. A few cases, however, have been reported. One of the most notable public reports was the Mexican use of oil risk markets around the Gulf War period. Developing country producers’ use of risk markets, however, may be small relative to its potential. For example, the World Bank reported in 1999 that less than 2 percent of the volume of commodity futures and options instruments could be attributed to developing countries. Even in oil contracts, developing countries were estimated to account for only 5 percent of open interest—that is, the number of outstanding contracts 4 (World Bank, 1999).
Governments of oil-exporting developed countries typically have less need to use hedging instruments because the share of total revenue accounted for by oil revenue is typically lower and because other sources of financing are more readily available. The state of Texas, however, is heavily reliant on oil revenue, and the government reportedly hedges this by executing collar spreads (buying put options financed by selling call options) to narrow the range within which this revenue stream fluctuates.
Oil market participants suggest that the major oil producers are generally aware of the commodity risk markets, and some have the expertise to make use of them.5 Some larger producers were also reported to have used the oil risk market, although their transactions were usually modest, transacted through third countries, and spread among a number of brokers. Market participants confirm, however, that oil producers usually make little use of oil risk markets.
Constraints on the Use of Contingent Markets
Several factors explain the relatively modest use of hedging. These include liquidity, creditworthiness, institutional capacity, the cost and use of capital, the perceived fairness of futures prices, basis risk, and the impact that large-scale hedging by producers might have on the market. These factors are discussed below.
Liquidity
An important constraint on hedging activity by producers is that the market may not be large enough. Hedging a substantial amount of future output may not be possible or may significantly distort market prices. Although there were grounds for such concerns until a few years ago, the situation nowadays is different, at least for the oil risk market.
Crude oil is the most actively traded commodity in the world. On an average day, NYMEX (New York Mercantile Exchange) trades about 200 million barrels of crude oil, equivalent to about three times daily global crude oil production and five times daily global crude oil exports. NYMEX crude oil open interest averages about a billion barrels, equivalent to about one month of global crude exports, two weeks of global crude production, or the annual crude oil export of Norway (the second largest oil exporter).
NYMEX has the largest trade in crude oil risk, followed closely by London’s International Petroleum Exchange (IPE). NYMEX futures and options extend seven years forward, but liquidity is concentrated at the short end. For both markets, about 75 percent of open interest is for six months or less forward, and about 5–10 percent for 24 months and beyond.
Information about over-the-counter (OTC) instruments is less readily available. The Bank for International Settlements, however, estimated that for end-1998, the notional amount of OTC nongold commodity contracts stood at $233 billion, almost all of which was petroleum- or metals-related. A major investment bank estimates that crude oil OTC volume is about twice as large as exchange-traded volume.
Market participants interviewed considered that they saw no substantial liquidity problems in hedging the annual output of midsized exporters from 6 to 18 months forward, if done gradually and using a range of instruments. Large producers could also increase revenue predictability and reduce volatility by hedging a fraction of their output. The annual output of top producers, however, could not be hedged without major market distortions, even if the market could address the moral hazard problems that would be involved.
The scope for substantial hedging beyond a few months forward for other nonrenewable commodities is limited. For metals, gold is the most active market, with some activity in the silver market. Transactions for gold can be made out to five years. Hedging length ranges from 6 to 12 months for base metals (copper, aluminum, lead, nickel, tin, and zinc). Copper and aluminum traded on the London Metal Exchange are the most liquid base metal risk markets, and the other base metals have very limited volume. While there are many swaps transacted for base metals, the liquidity and duration of the contracts are significantly less than in precious metals markets. Typical length of such transactions is three months to two years.
Creditworthiness
Credit constraints can be a significant factor in limiting large-scale access to certain types of risk instruments of less creditworthy countries. Long dated OTC transactions require the counterparty, often an international bank, to take the risk that the producing country or exporter might not pay if prices move against it over the contract period. Financial institutions have credit exposure limits to such countries, which may be a constraint in effecting such transactions.
Because of the lower credit exposure and the use of margins, shorter-dated exchange-traded transaction (for example, futures) are less hampered by these problems. The financial counterparty (for example, the futures broker), however, will still have to be assured about the country’s creditworthiness. The purchase of options, in contrast, does not involve any credit issues on the side of the purchaser.
Political Concerns
Politicians and the public may tend to view derivative transactions as inappropriate for governments. When hedging operations prove ex post profitable, the gains can be seen as speculative returns, and when they are ex post unprofitable, the losses can be seen as wasteful and irresponsible. Thus, although the success of hedging, particularly for purely speculative reasons, should be gauged by the lowering of risk and volatility associated with it, politicians may have little incentive to support hedging programs.
The incentive structure may also be flawed for those public sector officials who could undertake hedging operations. If prices fall, it is unlikely that the political authorities will blame the officials for not hedging, but if they fail to benefit from high prices, as could happen when futures are used extensively, politicians may well be critical. Hedging programs can, however, be structured to allow producers to gain from higher prices, for example, through the purchase of “put” options, which also protect financial positions against price declines.
Institutional Capacity
The personnel implications of implementing and monitoring hedging operations are significant. Risk management activities require considerable knowledge of financial instruments and an appropriate institutional framework within which to carry out hedging operations. Expertise is necessary to understand the risk structure of the company or public sector, identify appropriate risk management instruments, and engage in and supervise hedging transactions.
The institutional framework should also ensure adequate reporting, recording, monitoring, and evaluation mechanisms, and establish internal control procedures that can protect against speculative transactions and execution errors. Hedging operations are often complex, and without appropriately developed institutional capacity they can lead to less transparency and foster poor governance. Even highly sophisticated firms have been known to make large losses on hedging programs.
Cost and Use of Capital
Risk market transactions, especially for less creditworthy countries, often involve significant up-front premiums and margin calls. The use of futures requires the deposit of margins (usually 5–10 percent of the value of the underlying commodity). The purchase of options requires payment of a premium. Other commodity derivative instruments also require the use of capital for purchasing the instruments or for using collateral to cover performance risk. However, given the leverage that derivative transactions can allow (a small payment now can provide large gains or losses in the future), the cost of hedging may well be lower than obtaining similar levels of risk protection through other means, such as issuing debt.
“Fairness” of Futures Prices
Producers may consider that market prices for future production are unreasonably low. When futures prices are below spot prices, producers may reason that the spot price is just as likely to rise as to fall, and that they should not sell their future output cheaply. Similarly, when the spot price is historically low, futures prices may also be similarly low, unless futures prices are well above spot prices. Producers may then consider that the spot price is likely to rise to an equilibrium level in the future.
These views questionably assume that commodity producers are better at forecasting the future spot price than the market. For countries with substantial market-moving ability this may be true. But for other governments, it seems unlikely to be the case (Weiner, 1996). Also, as discussed in Section III, past prices may be poor indicators of future prices, and while the current spot price may be an indicator of future spot prices, futures market prices are somewhat more accurate and may be the best estimate available.
Basis Risk
The spot price of the specific type of oil a country produces may not be perfectly correlated with the financial instrument used to hedge it. This is known as basis risk. It implies that changes in the price received for a country’s nonrenewable resource may not move in lock step with the benchmark futures prices. This, however, does not necessarily remove the ability to hedge—it just reduces its efficiency.
A more specific form of basis risk is that the revenue the government receives may not be closely linked to the international oil price. For example, the government may receive royalties that are not related to the international oil price, so hedging against changes in the international oil price may not protect government revenue. In this case, however, tailor-made OTC instruments may be able to help, though these would be more costly and less liquid than more standard hedging instruments.
Market Impact
Forward sales by a large exporter may lead to an even greater negative market reaction. The possibility of a massive amount of forward supply and a sea change in the use of risk markets by large producers may push prices down further than justified purely by the size of any one transaction. If the producer is a member of a cartel, the effect (to ignore the intracartel considerations) may be even larger. As noted above, there are also moral hazard issues in dealing with producers who can affect market prices.
Hedging does not alter the revenue received directly from the sale of the commodity, but when this revenue is combined with the cash flow from the associated hedging operations, the combined commodity-related revenue stream can become less uncertain and volatile.
Different hedging strategies will lead to different levels of predictability and volatility. Using futures contracts exclusively will increase predictability but may not reduce volatility. Options, and other instruments, can also reduce volatility.
In the previous example, the government may consider that it could withstand an actual price of $20 per barrel in 2001 compared with a budget price of $26, but that prices below $15 a barrel would pose major problems. In this case, the government could buy today the option to sell in mid-2001 its output at $15 barrel.
Outstanding contracts comprise futures positions that have not been offset and option contracts that have not expired or been exercised
AS input for this paper, discussions were held with oil risk market participants in New York. Staff met with representatives from BNP Paribas, Fimat USA, Goldman Sachs, J. P. Morgan, and NYMEX.
Bibliography
Alier, Max, and Martin Kaufman, 1999, “Nonrenewable Resources: A Case for Persistent Fiscal Surpluses,” Working Paper 99/44 (Washington: International Monetary Fund).
Arrau, Patricio, and Stijn Claessens, 1992, “Commodity Stabilization Funds,” Worid Bank Policy Research Working Paper 835 (Washington: World Bank).
Auerbach, Alan J., Jagadeesh Gokhale, Laurence J. Kotlikoff, and Erling Steigum, Jr., 1993, “Generational Accounting in Norway: Is Norway Overconsuming Its Petroleum Wealth?” Ruth Pollak Working Papers Series on Economics, No. 24 (Boston: Boston University).
Basch, Miguel, and Eduardo Engel, 1993, “Temporary Shocks and Stabilization Mechanisms: The Chilean Case,” in External Shocks and Stabilization Mechanisms, Engel Eduardo Patricio and Meller (Washington: Inter-American Development Bank).
Cashin, Paul, and Hong, Liang, and C. John McDermott, 1999, “How Persistent Are Shocks to World Commodity Prices?” Working Paper 99/80 (Washington: International Monetary Fund).
Cashin, Paul, C. John McDermott, and Alasdair Scott, 1999, “Booms and Slumps in World Commodity Prices,” Working Paper 99/155 (Washington: International Monetary Fund).
Chalk, Nigel, 1998, “Fiscal Sustainability with Nonrenewable Resources,” Working Paper 98/26 (Washington: International Monetary Fund).
Claessens, Stijn, and Jonathan Coleman, 1991, “Hedging Commodity Price Risks in Papua New Guinea,” World Bank Policy, Research, and External Affairs Working Paper 749 (Washington: World Bank).
Claessens, Stijn, and Sweder van Wijnbergen, 1993, “1990 Mexico and Venezuela Recapture Clauses: An Application of Average Price Options,” Journal of Banking and Finance, Vol. 17 (June), pp. 733–45.
Claessens, Stijn, and Panos Varangis, 1994. “Oil Price Instability, Hedging, and an Oil Stabilization Fund: The Case of Venezuela,” World Bank Policy Research Working Paper 1290 (Washington: World Bank).
Corsetti, Giancarlo, and Nouriel Roubini, 1993, “The Design of Optimal Fiscal Rules for Europe After 1992,” in Adjustment and Growth in the European Monetary Union, Francisco Torres and Francisco Giavazzi (Cambridge: Cambridge University Press).
Engel, Eduardo, and Patricio Meller, eds., 1993, External Shocks and Stabilization Mechanisms (Washington: Inter-American Development Bank).
Engel, Eduardo, and Rodrigo Valdés, 2000, “Optimal Fiscal Strategy for Oil Exporting Countries,” Working Paper 00/118 (Washington: International Monetary Fund).
Fasano, Ugo, 2000, “Review of the Experience with Oil Stabilization and Savings Funds in Selected Countries,” Working Paper 00/112 (Washington: International Monetary Fund).
Gelb, Alan, and associates, 1988, Oil Windfalls: Blessing or Curse? (New York: Oxford University Press for the World Bank).
Hausmann, Ricardo, 1995, “Dealing with Negative Oil Shocks: The Venezuelan Experience in the Eighties,” Working Paper 307 (Washington: Inter-American Development Bank).
Hausmann, Ricardo, Andrew Powell, and Roberto Rigobón, 1993, “An Optimal Spending Rule Facing Oil Income Uncertainty (Venezuela),” in External Shocks and Stabilization Mechanisms,Eduardo Engel and Patricio Meller (Washington: Inter-American Development Bank).
International Monetary Fund, various years, World Economic Outlook (Washington).
Kopits, George, and Steven Symansky, 1998, Fiscal Policy Rules, Occasional Paper 162 (Washington: International Monetary Fund).
Kumar, Manmohan S., 1992, “Forecasting Accuracy of Crude Oil Futures Prices,” Staff Papers, International Monetary Fund, Vol. 39 (June), pp. 432
Liuksila, Claire, Alejandro Garcia, and Sheila Bassett, 1994, “Fiscal Policy Sustainabiiity in Oil-Producing Countries,” Working Paper 94/137 (Washington: International Monetary Fund),
Morauta, Hon. Sir Mekere. 2000, “Economic and Development Policies,” 2000 Budget of Papua New Guinea. Vol. 1 (Department of Finance and Treasury). Available via the Internet: http://www.treasury.gov.pg/treasury/treasury.nsf/pages/budget2000.
Potter, Barry H., and Jack Diamond, 1999, Guidelines for Public Expenditure Management International Monetary Fund).
Servén, Luis, and Andrés Solimano, eds., 1993, Striving for Growth after Adjustment: The Role of Capital Formation (Washington: World Bank).
Spilimbergo, Antonio, 1999, “Copper and the Chilean Economy, 1960–98.” Working Paper 99/57 (Washington: International Monetary Fund).
Steigum, Erling, Jr., and Øystein Thøgersen, 1995, “Petroleum Wealth, Debt Policy, and Intergenerational Welfare: The Case of Norway.” Journal for Policy Modeling, Vol. 17 (August), pp. 427–42.
Tersman, Gunnar, 1991, “Oil, National Wealth, and Current and Future Consumption Possibilities,” Working Paper 91/60 (Washington: International Monetary Fund).
United Nations Conference on Trade and Development, 1996, Price Risk Management in the Fuels Sector: A Manual (New York, Geneva: UNCTAD).
Varangis, Panos, Takamasa Akiyama, and Donald Mitchell, 1995, Managing Commodity Booms—and Busts (Washington: World Bank).
Varangis, Panos, and Don Larson, 1996, “Dealing with Commodity Price Uncertainty,” World Bank Policy Research Working Paper 1667 (Washington: World Bank).
Warrack, Allan A., and Russell R. Keddie, 2000, “Alberta Heritage Fund vs. Alaska Permanent Fund: A Comparative Analysis,” paper presented at the 39th Annual Meeting of the Western Regional Science Association, Kauai, Hawaii, February.
Weiner, Robert J., 1996, “Petroleum Fiscal Dependence: Revenue Forecasting and Oil Price Volatility,” George Washington University School of Business and Management Working Paper 96—44 (Washington: George Washington University).
Wickham, Peter, 1996, “Volatility of Oil Prices,” Working Paper 96/82 (Washington: International Monetary Fund).
World Bank, 1993, “Venezuela, Oil and Exchange Rates: Historical Experience and Policy Options,” World Bank Report No. 10481-VE (Washington).
World Bank, 1994, “Nigeria—Macroeconomic Risk Management: Issues and Options,” World Bank Report No. 11983-UNI (Washington).
World Bank, 1999, “Dealing with Commodity Price Volatility in Developing Countries: A Proposal for a Market-Based Approach,” Discussion Paper for the Round-table on Commodity Risk Management in Developing Countries, Washington, September.
See, for example. Gelb and associates (1988) and Engel and Meller (1993). Also, this paper does not discuss issues of fiscal federalism that may arise in nonrenewable resource-producing countries.
“Dutch disease” refers to the tendency for large resource revenues to appreciate the real exchange rate, which then damages the nonresource tradable sector.
It is usually easier to assess whether volume shocks are temporary or permanent.
If reversion to mean is slow, even if nonrenewable resource prices have a time-invariant long-run average, shocks will lend to seem permanent to countries with limited reserves of the nonrenewable resource.
See for example, Tersman (1991) and Engel and Valdés (2000).
Countries that have just discovered large nonrenewable resource endowments may he justified in consuming the initial revenue from the resource if it is below the estimated permanent income from the resource.
For example, a technological change may alter the equilibrium demand or supply of the resource, or both.
Simulations confirm the importance of the initial conditions for the results.
Research indicates that crude oil futures prices provide fore- casts that are, in general, superior to those obtained from alternative techniques for short-term horizons. For longer periods, their accuracy diminishes markedly; however, even for those horizons the futures forecasts are no worse, and are often better, compared with those obtained from alternative techniques (Kumar, 1992).
If the government did not raise its expenditure plans when the resource revenue rose, savings would increase automatically and there would be no need for a fund.
If the fund were allowed to spend directly or to lend, this would further impair any ability of the fund to generate financial savings.
For very large producers, hedging a substantial amount of future output may either not be possible or may distort market prices. Moreover, liquidity in contingent markets tends to be concentrated in the one- to two-year range. Credit constraints can also be an important factor in limiting large-scale excess to certain types of contingent markets by less creditworthy countries.
Fiscal rules may constrain expenditure or the deficit, or they may restrict the ability to borrow. As discussed in Kopits and Symansky (1998), fiscal rules can have both advantages and disadvantages.
These risks would likely be increased when earmarking lakes place for off-budget expenditures.
These arguments, however, may not apply in the case of perfect capital mobility and highly developed domestic financial markets, and when the operations of the fund are small relative to the size of the domestic financial market.
Performance evaluation should be distinguished from financial compliance assessments: the latter aims to ensure that all funds are accounted for, while the performance evaluation aims to assess whether funds were used in the best possible way and adhered to investment guidelines.
A forward-looking government would alter spending on the basis of anticipated revenue. Thus, even if the revenue from higher resource export earnings does not reach the budget for some time, government spending could respond immediately. The endogeneity could arise if profit transfers from the slate resource enterprise are determined at least in part by the level of expenditure.
Papua New Guinea had an NRF throughout the sample period and thus could not be included
See also Fasano (2000).
The overall budget balance is calculated with oil investments made on behalf of the State Direct Financial Interest in the oil.
Information on the fund’s assets is not publicly available.
The Oil Fund receives oil revenues equivalent to the market value of a fixed volume of oil production. In 1998, the government borrowed from the fund to finance the budget: this was repaid in 1999.
Partly in reflection of the desire to insulate the fund from spending pressures, provision of public information on the fund’s assets is prohibited by law.
The government concluded in the 2000 budget that the performance of the MRSF had not been as originally envisaged, and that the stabilization function of the legislation had proven ineffective in the face of the fiscal decision makers’ willingness to issue excessive volumes of government debt (Morauta, 2000).
OCCASIONAL PAPERS: Recent Occasional Papers of the International Monetary Fund
205. Stabilization and Savings Funds for Nonrenewable Resources; Experience and Fiscal Policy Implications, by Jeffrey Davis, Rolando Ossowski, James Daniel, and Steven Barnett. 2001.
204. Monetary Union in West Africa (ECOWAS): Is It Desirable and How Could It Be Achieved? by Paul Masson and Catherine Pattillo. 2001.
203. Modern Banking and OTC Derivatives Markets: The Transformation of Global Finance and Its Implications for Systemic Risk, by Garry J. Schinasi, R. Sean Craig, Burkhard Drees, and Charles Kramer. 2000.
202. Adopting Inflation Targeting: Practical Issues for Emerging Market Countries, by Andrea Schaechter, Mark R. Stone, and Mark Zelmer. 2000.
201. Developments and Challenges in the Caribbean Region, by Samuel Itam, Simon Cueva, Erik Lundback, Janet Stotsky, and Stephen Tokarick. 2000.
200. Pension Reform in the Baltics: Issues and Prospects, by Jerald Schiff, Niko Hobdari, Axel Schimmelpfennig, and Roman Zytek. 2000.
199. Ghana: Economic Development in a Democratic Environment, by Sé Ygio Pereira Leite, Anthony Pellechio, Luisa Zanforlin, Girma Begashaw, Stefania Fabrizio, and Joachim Harnack, 2000.
198. Setting Up Treasuries in the Baltics, Russia, and Other Countries of the Former Soviet Union: An Assessment of IMF Technical Assistance, by Barry H. Potter and Jack Diamond. 2000.
197. Deposit Insurance: Actual and Good Practices, by Gillian G.H. Garcia. 2000.
196. Trade and Trade Policies in Eastern and Southern Africa, by a staff team led by Arvind Subramanian, with Enrique Gelbard, Richard Harmsen, Katrin Elborgh-Woytek, and Piroska Nagy. 2000.
195. The Eastern Caribbean Currency Union—Institutions, Performance, and Policy Issues, by Frits van Beek, José Roberto Rosales, Mayra Zermeño, Ruby Randall, and Jorge Shepherd. 2000.
194. Fiscal and Macroeconomic Impact of Privatization, by Jeffrey Davis, Rolando Ossowski, Thomas Richardson, and Steven Barnett. 2000.
193. Exchange Rate Regimes in an Increasingly Integrated World Economy, by Michael Mussa, Paul Masson, Alexander Swoboda, Esteban Jadresic, Paolo Mauro, and Andy Berg, 2000.
192. Macroprudential Indicators of Financial System Soundness, by a staff team led by Owen Evans, Alfredo M. Leone, Mahinder Gill, and Paul Hilbers. 2000.
191. Social Issues in IMF-Supported Programs, by Sanjeev Gupta, Louis Dicks-Mireaux, Ritha Khemani, Calvin McDonald, and Marijn Verhoeven. 2000.
190. Capital Controls: Country Experiences with Their Use and Liberalization, by Akira Ariyoshi, Karl Habermeier, Bernard Laurens, Inci Ötker-Robe, Jorge Ivan Canales Kriljenko, and Andrei Kirilenko. 2000.
189. Current Account and External Sustainability in the Baltics, Russia, and Other Countries of the Former Soviet Union, by Donal McGettigan. 2000.
188. Financial Sector Crisis and Restructuring: Lessons from Asia, by Carl-Johan Lindgren, Tomas J.T. Baliño, Charles Enoch, Anne-Marie Guide, Marc Quintyn, and Leslie Teo. 1999.
187. Philippines: Toward Sustainable and Rapid Growth, Recent Developments and the Agenda Ahead, by Markus Rodlauer, Prakash Loungani, Vivek Arora, Charalambos Christofides, Enrique G. De la Piedra, Piyabha Kongsamut, Kristina Kostial, Victoria Summers, and Athanasios Vamvakidis. 2000.
186. Anticipating Balance of Payments Crises: The Role of Early Warning Systems, by Andrew Berg, Eduardo Borensztein, Gian Maria Milesi-Ferretti, and Catherine Pattillo. 1999.
185. Oman Beyond the Oil Horizon: Policies Toward Sustainable Growth, edited by Ahsan Mansur and Volker Treichel. 1999.
184. Growth Experience in Transition Countries, 1990-98, by Oleh Havrylyshyn, Thomas Wolf, Julian Beren-gaut, Marta Castello-Branco, Ron van Rooden, and Valerie Mercer-Blackman. 1999.
183. Economic Reforms in Kazakhstan, Kyrgyz Republic, Tajikistan, Turkmenistan, and Uzbekistan, by Emine Gürgen, Harry Snoek, Jon Craig, Jimmy McHugh, Ivailo Izvorski, and Ron van Rooden. 1999.
182. Tax Reform in the Baltics, Russia, and Other Countries of the Former Soviet Union, by a staff team led by Liam Ebrill and Oleh Havrylyshyn. 1999.
181. The Netherlands: Transforming a Market Economy, by C. Maxwell Watson, Bas B. Bakker, Jan Kees Martijn, and Ioannis Halikias. 1999.
180. Revenue Implications of Trade Liberalization, by Liam Ebrill, Janet Stotsky, and Reint Gropp. 1999.
179. Disinflation in Transition: 1993-97, by Carlo Cottarelli and Peter Doyle. 1999.
178. IMF-Supported Programs in Indonesia, Korea, and Thailand: A Preliminary Assessment, by Timothy Lane, Atish Ghosh, Javier Hamann, Steven Phillips, Marianne Schulze-Ghattas, and Tsidi Tsikata. 1999.
177. Perspectives on Regional Unemployment in Europe, by Paolo Mauro, Eswar Prasad, and Antonio Spilimbergo. 1999.
176. Back to the Future: Postwar Reconstruction and Stabilization in Lebanon, edited by Sena Eken and Thomas Helbling. 1999.
175. Macroeconomic Developments in the Baltics, Russia, and Other Countries of the Former Soviet Union, 1992-97, by Luis M. Valdivieso. 1998.
174. Impact of EMU on Selected Non—European Union Countries, by R. Feldman, K. Nashashibi, R. Nord, P. Allum, D. Desmelle, K. Enders, R. Kahn. and H. Temprano-Arroyo. 1998.
173. The Baltic Countries: From Economic Stabilization to EU Accession, by Julian Berengaut, Augusto Lopez-Claros, Franchise Le Gall, Dennis Jones, Richard Stern, Ann-Margret Westin, Effie Psalida, Pietro Garibaldi. 1998.
172. Capital Account Liberalization: Theoretical and Practical Aspects, by a staff team led by Barry Eichen-green and Michael Mussa, with Giovanni Dell’Ariccia, Enrica Detragiache, Gian Maria Milesi-Ferretti, and Andrew Tweedie. 1998.
171. Monetary Policy in Dollarized Economies, by Tomas Balino, Adam Bennett, and Eduardo Borensztein. 1998.
170. The West African Economic and Monetary Union: Recent Developments and Policy Issues, by a staff team led by Ernesto Herná ndez-Catá and comprising Christian A. Francois, Paul Masson, Pascal Bouvier, Patrick Peroz, Dominique Desruelle, and Athanasios Vamvakidis. 1998.
169. Financial Sector Development in Sub-Saharan African Countries, by Hassanali Mehran, Piero Ugolini, Jean Phillipe Briffaux, George Iden, Tonny Lybek, Stephen Swaray, and Peter Hayward. 1998.
168. Exit Strategies: Policy Options for Countries Seeking Greater Exchange Rate Flexibility, by a staff team led by Barry Eichengreen and Paul Masson with Hugh Bredenkamp, Barry Johnston, Javier Hamann, Esteban Jadresic, and Inci Ötker. 1998.
167. Exchange Rate Assessment: Extensions of the Macroeconomic Balance Approach, edited by Peter lsard and Hamid Faruqee. 1998
166. Hedge Funds and Financial Market Dynamics, by a staff team led by Barry Eichengreen and Donald Mathieson with Bankim Chadha, Anne Jansen, Laura Kodres, and Sunil Sharma. 1998.
165. Algeria: Stabilization and Transition to the Market, by Karim Nashashibi. Patricia Alonso-Gamo, Stefania Bazzoni, Alain FéMer, Nicole Laframboise, and Sebastian Paris Horvitz. 1998.
164. MULTIMOD Mark III: The Core Dynamic and Steady-State Model, by Douglas Laxton, Peter lsard, Hamid Faruqee, Eswar Prasad, and Bart Turtelboom. 1998.
163. Egypt: Beyond Stabilization, Toward a Dynamic Market Economy, by a staff team led by Howard Handy. 1998.
162. Fiscal Policy Rules, by George Kopits and Steven Symansky. 1998.
161. The Nordic Banking Crises: Pitfalls in Financial Liberalization? by Burkhard Drees and Ceyla Pazarbaşioğtu. 1998.
Note: For information on the title and availability of Occasional Papers not listed, please consult the IMF Publications Catalog or contact IMF Publication Services.