Abstract
This Selected Issues and Statistical Appendix paper on Gabon reviews management of oil revenues, competitiveness, and growth. The nature of Gabon’s problems has not changed during the past 15 years. The need to diversify the economy and the export base; control fiscal expenditure and the wage bill; carefully assess capital expenditure; and reform public sector enterprises are the challenges that the Gabonese need to be prepared to implement adequately. Gabon faces huge medium-term fiscal constraints imposed by the expected steady decline in oil production and its depletion.
II. Managing Oil Revenues in Gabon—Is There a Role for the Fund for Future Generations?1
A. Introduction: Two Views on Oil Revenues
1. A country that is blessed with abundant natural resources—notably oil—faces several macroeconomic challenges. Export receipts, as well as fiscal revenues from resource extraction are often highly volatile, requiring the government to plan budgetary expenditures carefully. This task is complicated by the need to distinguish temporary shocks from longer-term or permanent ones. Even when the path of expenditures is relatively well known (such as with long-term investment projects and recurrent expenditures), if future oil revenues are the main source for paying for these engagements, then a certain degree of prudence is still required. Otherwise, governments run the risk of ending up with an unsustainable debt.
2. Governments must also take into consideration the dynamic effects that revenues from oil have on the rest of the economy. For example, rapid expansion of the “enclave” oil sector and large inflows of foreign exchange may put upward pressure on the currency and prices, hampering competitiveness and the development of other sectors (see Chapter III). Also, large oil revenues may hamper the development of a broad-based tax system and may weaken discipline in overall budgetary management and governance.
3. Finally, governments that rely on revenues from natural resource extraction must address the fact that those revenues are exhaustible. The finite nature of natural resource extraction opens the question of the intertemporal allocation of the economic benefits from such resources. One view is to interpret revenues from extraction as current income, in which case the intertemporal allocation of extraction revenues boils down to a standard consumption and savings decision. Due to the volatility in receipts, the decision to save may be based on precautionary motives, resulting in attempts to smooth the impact of income volatility on consumption. Or, the savings decision may be motivated by a bequest motive that leaves assets for future generations. If the savings are put to productive use, then future generations will inherit a higher capital stock and would be better off (even with low yields on those savings) than if all current revenues were consumed immediately.
4. An alternative view is to consider natural resource revenues as a transformation of natural resource wealth into financial wealth, rather than an income stream. Total wealth is then the sum of natural resource wealth (the present value of current and all future extraction revenues) and financial wealth.2 In managing natural resource wealth, each generation must then answer two basic questions. First, how much of the natural resource reserves will be extracted each period? In principle, the answer to this question depends both on technological constraints and on the risk-return characteristics of each type of asset (natural and financial) in the overall wealth portfolio. For example, when world oil prices are temporarily “high,” it may be desirable to speed up oil extraction by operating marginal fields. The second question is how much should be consumed in each period. If all the extraction revenues are immediately consumed, then the present generation effectively pays for current consumption by reducing assets and does so at the expense of all future generations. The other extreme would be to invest all extraction revenues in financial assets, leaving everything for future generations. Clearly, there exist an infinite number of intermediate choices that provide a more evenly distributed outcome across generations. The “optimal” intergenerational distribution depends on society’s intertemporal welfare function and is hard to determine.3 One choice is to ensure intergenerational equity by letting only the permanent income from wealth accrue to each generation. This permanent income may be thought of as an annuity that allows each generation to consume without depleting wealth for future generations. In a society with population growth, the annuity value would be calculated as the permanent income from a constant real per capita stock of wealth of all current and future members of society. Alternatively, countries with a relatively low capital stock might prefer to spend more on investment, thereby helping them achieve their steady-state capital stock faster.4
5. The intertemporal choices become more complicated when a society has accumulated debt. In that case, the extraction revenues could be used for immediate consumption, for building up gross financial wealth, and for paying off debt. The optimal composition of the nation’s net wealth portfolio divides this portfolio into natural resource wealth, gross financial wealth, and debt. In principle, risk-return characteristics of each of the three components should guide this decision. If the interest rate on debt is higher than the expected return on financial assets, then it may be desirable to pay off debt first before building up financial assets. However, it may be optimal to keep some financial assets for precautionary reasons, even if their expected return is lower. If access to credit is constrained during bad times, such a stock of financial assets would allow smoothing of consumption expenditures and also guarantee the timeliness of future debt service payments in light of volatile extraction revenues.
6. In the context of ensuring intergenerational equity, the decision to pay off debt with oil revenues may depend on the origins of the debt. If the debt was accumulated to finance investments in productive activities, and both the current and future generations stand to benefit from a higher capital stock, then it can be argued that all generations share in paying off the debt. In that case, it would be justified to reduce the debt burden through the use of oil revenues. A moral argument may suggest a different approach if the debt was accumulated primarily to pay for consumption of an impatient present generation, of if the debt financed largely unproductive and wasteful investment spending. In that case, a more just outcome might be to save more oil revenues for future generations and pay off the debt mainly through non-oil primary surpluses (that is, less current spending, increased non-oil revenues, or both).
7. This paper focuses on the question of how useful it is to put some of the Gabonese oil revenues aside in a Fund for Future Generations, with a focus on the institutional arrangements of the Fund as outlined in its founding legal provisions. A full assessment of the intergenerational aspects of the Fund for Future Generations requires making assumptions about the current and future path of government revenues and expenditures. The reason is that, for example, simply setting aside part of oil revenues while financing non-oil deficits with expensive debt would not leave future generations better off, as net wealth declines. Thus, a complete analysis would impose fiscal rules on the non-oil fiscal deficit, and then derive the required fiscal adjustment for given contributions to the Fund. However, such an extensive exercise is beyond the scope of this paper. Instead, a more limited approach is taken by focusing exclusively on the contribution of the Fund to consumption of present and future generations and comparing the outcomes of different scenarios of remuneration. This is justified as long as the contributions are made from any remaining oil revenues, after first financing the non-oil deficit. Once the non-oil deficit is financed, the remaining choices are those of portfolio allocation. In this paper, the opportunity cost of contributing to the Fund (building up other financial assets or paying down debt) is captured by the discount rate.
8. The rest of the paper is organized as follows. Sections B and C give examples of natural resource funds from various countries and describe the modalities of the Gabonese Fund for Future Generations in detail. In Section D the assumptions are developed and the results of various scenarios of the potential effects on present and future generations are presented. Finally, Section E sums up the findings and concludes.
B. Natural Resource Funds
9. Several countries with natural resource revenues have set up funds in which some or all such revenues are deposited. The objectives, operational rules, transparency requirements and the actual performance of such funds vary widely; a detailed description of the modalities is beyond the scope of this paper.5 One example is the Norwegian State Petroleum Fund, which is a separate government account where oil revenues are deposited. In return, the fund finances the non-oil fiscal deficit through a transfer to the budget. The Norwegian oil fund has been credited with improving the transparency of the budgetary process, and makes clearer the intertemporal choices facing the government, which sets explicit targets for the non-oil budget deficit.
10. In other cases, the natural resource fund has mainly a stabilization objective, such as with the Chilean Copper Stabilization Fund, the Mineral Resources Stabilization Fund in Papua New Guinea (abolished in 2001), and the Macroeconomic Stabilization Fund in Venezuela. The decision to contribute and withdraw may depend on price triggers that could be rules-based or left up to the discretion of policymakers.
11. Finally, in some cases the goal of the fund is to be a store of value for future generations. Examples of this class of “savings” funds include the Reserve Fund for Future Generations in Kuwait and the Alaska Permanent Fund set up by the State of Alaska. In both cases, contributions are made independently of world oil market developments or of the budgetary situation of the government.
C. The Fund for Future Generations
12. Gabon is a mature oil producing economy. Barring any significant new discoveries, oil production is expected to decline substantially during the next decade. The intergenerational dimension of the projected decline in oil production is given higher urgency by the low degree of diversification of the economy, the reliance of the government budget on oil revenues, and the large stock of outstanding debt.
13. To safeguard some of the oil revenues for future generations, a law was passed in July 1998 that created a Fund for Future Generations (Fonds pour les Générations Futures, FGF). The FGF is conceived as a perpetual reserve fund with a minimum capital of CFAF 500 billion. The 1998 law explicitly bars the use of the resources in the FGF to finance government spending.
14. The law describes the rules for contributions into the FGF. During a first phase, which applies up to the minimum capital is reached, the following contribution rules apply:
10 percent of projected revenues, using a baseline projection that is determined in the annual budget law;
50 percent of oil revenues exceeding the baseline projection contained in the budget law (that is, half of the “windfall” revenues);
reinvestment of all the interest income from the FGF.
15. Once the minimum capital is reached, contributions would be made as follows:
all the oil revenues exceeding the baseline projection set out in the budget law (that is, all of the windfall revenues);
reinvestment of one fourth of the interest income from the FGF (the remaining three quarters of interest income is paid into the general resources of the government).
16. The FGF is held in a special account at the regional central bank (BEAC). While originally not remunerated, since January 2005 the BEAC pays interest on the outstanding balance at a rate of 1.7 percent per annum.
17. In practice, the government has not made the contributions to the FGF as outlined in the 1998 law, and at end-2004, the outstanding balance of the FGF stood at CFAF 55 billion (1.43 percent of GDP).
D. Economic Analysis of the Fund for Future Generations
Main assumptions
18. To better understand the potential role of the FGF, different scenarios about the future path of production, prices and contribution and investment rules are analyzed. Table 1 summarizes the assumptions about production and prices. For most scenarios the 2005-10 oil projections of the Gabonese Direction Générale des Hydrocarbures (DGH) are extended to 2030. Under the DGH projections, oil extraction will decline from 98.9 million barrels in 2005 to 54.8 million barrels in 2010. These projections are extended by applying average decline rates per field, and by also assuming that the oil fields will be exhausted by 2030.
Gabon: Projected Oil Production and Prices
Gabon: Projected Oil Production and Prices
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2020 | 2030 | |
|---|---|---|---|---|---|---|---|---|
| Oil production (Millions of barrels) (DGH) | 98.9 | 88.7 | 80.2 | 70.1 | 62.7 | 54.8 | 17.4 | 6.2 |
| Exchange rate (CFA francs/U.S. dollar) | 488.4 | 488.3 | 488.0 | 487.8 | 487.6 | 487.4 | 487.4 | 487.4 |
| Brent price (U.S. dollars per barrel) | 48.00 | 45.25 | 42.50 | 41.25 | 40.50 | 40.00 | 35.00 | 35.00 |
Gabon: Projected Oil Production and Prices
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2020 | 2030 | |
|---|---|---|---|---|---|---|---|---|
| Oil production (Millions of barrels) (DGH) | 98.9 | 88.7 | 80.2 | 70.1 | 62.7 | 54.8 | 17.4 | 6.2 |
| Exchange rate (CFA francs/U.S. dollar) | 488.4 | 488.3 | 488.0 | 487.8 | 487.6 | 487.4 | 487.4 | 487.4 |
| Brent price (U.S. dollars per barrel) | 48.00 | 45.25 | 42.50 | 41.25 | 40.50 | 40.00 | 35.00 | 35.00 |
19. The assumptions for exchange rates and Brent oil prices up to 2010 are based on the March 2005 WEO projections. After 2010 the euro/dollar exchange rate is assumed to remain unchanged from 2010, which translates in a CFA franc/U.S. dollar rate of 487.4. The benchmark Brent price of oil is projected to fall from US$48 per barrel in 2005 to US$40 in 2010. After 2010 the Brent price is assumed to converge gradually to US$35 per barrel by 2020, and remain at that level thereafter.
20. To calculate the oil revenues the authorities’ oil model is used and extended to 2030. This model takes into account the specific contractual arrangements of each oil field (concession contracts and production sharing agreements) and also makes assumptions about the price of each type of Gabonese oil relative to the Brent price benchmark.
21. To calculate the net present value of future revenue streams, a discount rate of 3.29 percent is used, which is the WEO average projected euro Libor rate over the period 2005-10. Inflation is projected to remain at 1 percent, resulting in a real discount rate of 2.27 percent. All the scenarios are calculated and reported in constant 2005 prices. The population is assumed to grow at 2 percent, which is somewhat lower than the current growth rate of the population.
22. To calculate the contributions to the FGF, the parameters set in the 1998 law are used. This requires an assumption about the projected windfall oil revenues for the government, that is, an assumption of the actual oil revenues relative to those projected in the budget. Since the authorities typically project oil revenues based on conservative assumptions, the windfall is calculated by assuming a reference Brent price of US$30 per barrel. Given the WEO projections for the Brent price of oil, this implies substantial windfall revenues in the early projection years, converging to a US$5 windfall per barrel after 2010.
Main scenario: contributions start in 2005
23. Three scenarios for the intertemporal effects of the FGF are developed, applying the rules set in the 1998 law from 2005 onwards. The first scenario is a benchmark in which all oil revenues are immediately consumed and nothing is saved. Given the projected production and the assumptions about the Brent price of oil, this implies that there will be no revenues left after 2030. In the second scenario the FGF is introduced, with the assumption that all contributions are made in line with the 1998 law, starting in 2005. In particular, the FGF receives an interest rate of 1.7 percent. Finally, the third scenario follows the same contribution guidelines but assumes that the FGF is remunerated at the euro Libor rate of 3.29 percent (the same rate as the nominal discount rate).
24. Table 2 summarizes the results from the three scenarios. The first column gives the present discounted value of all future flows. The table shows total oil revenues, the value of the FGF, total and per capita at different years up to 2030.
Gabon: Value of FGF, Total and Per Capita Consumption
(Constant 2005 CFA franc)
Gabon: Value of FGF, Total and Per Capita Consumption
(Constant 2005 CFA franc)
| Present Disc. Value (2005 CFAF) | 2005 | 2006 | 2010 | 2015 | 2020 | 2025 | 2030 | ||
|---|---|---|---|---|---|---|---|---|---|
| Oil Revenues (billion of CFA francs) | 4,401.3 | 731.7 | 617.6 | 315.7 | 156.9 | 77.0 | 41.9 | 23.2 | |
| Value FGF (billion of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | |
| Scenario 2 (FGF at 1.7 percent) | 556.2 | 253.2 | 416.5 | 883.7 | 1,157.9 | 1,239.0 | 1,254.9 | 1,248.6 | |
| Scenario 3 (FGF at 3.29 percent) | 1,247.8 | 254.1 | 421.4 | 903.1 | 1,197.1 | 1,301.3 | 1,341.2 | 1,358.7 | |
| Consumption (billion of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 4,401.3 | 731.7 | 617.6 | 315.7 | 156.9 | 77.0 | 41.9 | 23.2 | |
| Scenario 2 (FGF at 1.7 percent) | 3,766.4 | 534.4 | 456.1 | 229.7 | 130.7 | 78.1 | 49.7 | 34.3 | |
| Scenario 3 (FGF at 3.29 percent) | 4,458.1 | 534.4 | 456.1 | 239.4 | 144.8 | 94.0 | 66.5 | 51.6 | |
| Per capita consumption | |||||||||
| (Thousand of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 562.8 | 465.8 | 219.9 | 99.0 | 44.0 | 21.7 | 10.9 | ||
| Scenario 2 (FGF at 1.7 percent) | 411.0 | 343.9 | 160.0 | 82.5 | 44.6 | 25.7 | 16.1 | ||
| Scenario 3 (FGF at 3.29 percent) | 411.0 | 343.9 | 166.8 | 91.4 | 53.7 | 34.4 | 24.2 | ||
Gabon: Value of FGF, Total and Per Capita Consumption
(Constant 2005 CFA franc)
| Present Disc. Value (2005 CFAF) | 2005 | 2006 | 2010 | 2015 | 2020 | 2025 | 2030 | ||
|---|---|---|---|---|---|---|---|---|---|
| Oil Revenues (billion of CFA francs) | 4,401.3 | 731.7 | 617.6 | 315.7 | 156.9 | 77.0 | 41.9 | 23.2 | |
| Value FGF (billion of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | |
| Scenario 2 (FGF at 1.7 percent) | 556.2 | 253.2 | 416.5 | 883.7 | 1,157.9 | 1,239.0 | 1,254.9 | 1,248.6 | |
| Scenario 3 (FGF at 3.29 percent) | 1,247.8 | 254.1 | 421.4 | 903.1 | 1,197.1 | 1,301.3 | 1,341.2 | 1,358.7 | |
| Consumption (billion of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 4,401.3 | 731.7 | 617.6 | 315.7 | 156.9 | 77.0 | 41.9 | 23.2 | |
| Scenario 2 (FGF at 1.7 percent) | 3,766.4 | 534.4 | 456.1 | 229.7 | 130.7 | 78.1 | 49.7 | 34.3 | |
| Scenario 3 (FGF at 3.29 percent) | 4,458.1 | 534.4 | 456.1 | 239.4 | 144.8 | 94.0 | 66.5 | 51.6 | |
| Per capita consumption | |||||||||
| (Thousand of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 562.8 | 465.8 | 219.9 | 99.0 | 44.0 | 21.7 | 10.9 | ||
| Scenario 2 (FGF at 1.7 percent) | 411.0 | 343.9 | 160.0 | 82.5 | 44.6 | 25.7 | 16.1 | ||
| Scenario 3 (FGF at 3.29 percent) | 411.0 | 343.9 | 166.8 | 91.4 | 53.7 | 34.4 | 24.2 | ||
25. The present discounted value of all future oil revenues is calculated at CFAF 4,401.3 billion. This is the total amount of oil wealth, in 2005 CFA francs, that can be shared among the present and all future generations. In the first scenario, this oil wealth is consumed at the same rate as oil is extracted. Under this scenario, total consumption is projected to fall from CFAF 731.7 billion in 2005 to CFAF 23.2 billion in 2030, and is zero forever after. In per capita terms, consumption falls from CFAF 562,800 in 2005 to CFAF 10,900 by 2030.
26. This first “hand to mouth” scenario is used as a reference for comparison with the second and third scenarios, in which contributions to the FGF are introduced. In the second scenario, with the FGF remunerated at a nominal rate of 1.7 percent, the total value of the FGF is calculated at CFAF 556.2 billion. The effect of the FGF is to reduce short-term consumption (as oil revenues are set aside while the FGF is built up), but to increase future consumption once it pays out interest into the general budget. Compared to the benchmark scenario, total annual consumption spending in 2005 is reduced from CFAF 731.7 billion to By 2030 though, consumption spending increases from CFAF 23.2 billion in the “hand to mouth” scenario to CFAF 34.3. The difference is entirely due to the extra consumption spending that comes from the FGF resources. In addition, by end-2030 the balances in the fund have grown to CFAF 1,248.6 billion; this amount is available for spending after 2030 (in contrast, under the first scenario, nothing is available for spending after 2030). In per capita terms, savings into the FGF reduces 2005 consumption from CFAF 562,800 to CFAF 411,000 (a reduction of 26.9 percent), but by 2030 the savings fund allows per capita consumption to be 47.7 percent higher than under the benchmark scenario (CFAF 16,100 instead of CFAF 10,900). The savings into the FGF allow per capita consumption to be higher than the no-savings scenario from 2020 onwards.
27. From the comparison between the benchmark (no-savings) scenario and the (low-remuneration) second scenario, it is found that the FGF helps preserve wealth for future generations, leaving them significantly better off compared to the situation in which no savings are set aside. Further, consumption is higher even before oil revenues dry up.
28. Next, the implications of the specific remuneration rate of the fund are assessed. To do this, a third scenario is developed in which the FGF balances earn a projected euro Libor rate of 3.29 percent instead of the relatively low rate of 1.7 percent from the second scenario. The results form the third scenario are also reported in Table 2. Just like in the other two scenarios, total consumption drops dramatically between 2005 and 2030, in this case from CFAF 534.4 billion to CFAF 51.6 billion respectively. In per capita terms this translates into a reduction from CFAF 411,000 in 2005 to CFAF 24.2 in 2030. However, when these magnitudes are compared to those from the low-remuneration scenario, it is found that future generations are left substantially better off when the fund earns the higher market return. For example, per capita consumption is 10.8 percent higher in 2015 (CFAF 91,400 versus CFAF 82,500), and by 2030 per capita consumption is 50 percent higher (and more than double than in the no-savings scenario). The effect of the differences in remuneration also shows up in the capitalization value of the FGF: it is calculated at CFAF 556.2 billion in the low-remuneration scenario, versus CFAF 1,247.8 billion when a market interest is earned. By the time when oil revenues dry up completely (in 2030), future generations will inherit a fund with a total value of CFAF 1,358.7 billion, almost 9 percent more than the CFAF 1,248.6 billion that accumulates under the 1.7 percent remuneration scenario.
29. Further evidence of the impact of using different remuneration rates is found when the present discounted values of total consumption are compared. For the third scenario, this value is calculated at CFAF 4,458.1 billion, which is equal to the sum of the outstanding value of the fund at end-2004 and the present discounted value of all future oil revenues. However, in the low-remuneration scenario, the present discounted value of total consumption is only CFAF 3,766.4 billion. That is, while redistributing consumption from present generations to future generations, the low-remuneration scenario achieves this at the cost of leaving all generations worse off by a combined amount of CFAF 691.7 billion. This is entirely due the fact that the interest rate on the outstanding balances is less than the discount rate, which reduces the value of all savings.6
30. Although both scenarios in which oil revenues are saved into the FGF increase consumption for future generations, neither achieves perfect intergenerational equity. To achieve such equity would require that only the permanent income from oil wealth is made available for annual consumption spending. With a capitalization value of oil wealth equal to CFAF 4,401.3 billion, total annual permanent income would be equal to CFAF 99.8 billion. However, with population growth this annuity amount would still imply declining per capita consumption spending. For that reason the per capita annuity APC is also calculated, which is the permanent income from a constant per capita wealth for each current and future individual. This amount is calculated as follows:
APC = (W2005/P2005)*(r - g)/(1 + r)
31. In the expression, W2005 is the present discounted value of all oil wealth, P2005 is the population in 2005, r is the discount rate, and finally g is the growth rate of population. Applying the assumptions gives a per capita annuity of CFAF 8,850. It is clear that both the total annual annuity and the per capita annuity are much lower than short-term and medium-term consumption spending of each of the three scenarios reported in Table 2. For example, compared to the no-savings scenario, for 2005 the constant annuity would imply that CFAF 631.9 billion of oil revenues should be saved; the per capita annuity would require a reduction in per capita consumption of CFAF 553,950. For the per capita annuity, the low amount is due mainly because the real discount rate and the population growth rate are close (2.27 percent and 2.0 percent, respectively), which requires that most of the oil revenues must be saved in order to preserve sufficient per capita wealth for all future generations.7
32. The results from the intergenerational equity calculations for each of the three scenarios are reported in Table 3. The most striking observation is that the annuities from the second scenario are smaller than those from both the benchmark scenario and the high-remuneration scenario. For the total annual annuity, this amount is CFAF 85.4 billion versus CFAF 99.8 billion and CFAF 101.1 billion, respectively. Again, the low remuneration of the FGF leaves a smaller permanent income available for all generations, confirming the earlier findings.
Gabon: Intergenerational Equity: Annuity Values (Permanent Income) Compared to FGF Scenarios
Unit expressed in billion of 2005 CFA franc.
Unit expressed in thousand of 2005 CFA franc.
Gabon: Intergenerational Equity: Annuity Values (Permanent Income) Compared to FGF Scenarios
| 2005 | 2006 | 2010 | 2015 | 2020 | 2025 | 2030 | ||
|---|---|---|---|---|---|---|---|---|
| Constant total spending | ||||||||
| Scenario 1 (no FGF) 1/ | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | |
| Implied per capita spending 2/ | 76.8 | 75.3 | 69.5 | 63.0 | 57.0 | 51.7 | 46.8 | |
| Scenario 2 (FGF at 1.7 percent) 1/ | 85.4 | 85.4 | 85.4 | 85.4 | 85.4 | 85.4 | 85.4 | |
| Implied per capita spending 2/ | 65.7 | 64.4 | 59.5 | 53.9 | 48.8 | 44.2 | 40.0 | |
| Scenario 3 (FGF at 3.29 percent) 1/ | 101.1 | 101.1 | 101.1 | 101.1 | 101.1 | 101.1 | 101.1 | |
| Implied per capita spending | 77.8 | 76.2 | 70.4 | 63.8 | 57.8 | 52.3 | 47.4 | |
| Constant per capita spending | ||||||||
| Scenario 1 (no FGF) 2/ | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | |
| Implied total spending 1/ | 11.5 | 11.7 | 12.7 | 14.0 | 15.5 | 17.1 | 18.9 | |
| Scenario 2 (FGF at 1.7 percent) 2/ | 7.75 | 7.75 | 7.75 | 7.75 | 7.75 | 7.75 | 7.75 | |
| Implied total spending 1/ | 9.8 | 10.0 | 10.9 | 12.0 | 13.3 | 14.6 | 16.2 | |
| Scenario 3 (FGF at 3.29 percent) 2/ | 8.96 | 8.96 | 8.96 | 8.96 | 8.96 | 8.96 | 8.96 | |
| Implied total spending 1/ | 11.7 | 11.9 | 12.9 | 14.2 | 15.7 | 17.3 | 19.1 | |
Unit expressed in billion of 2005 CFA franc.
Unit expressed in thousand of 2005 CFA franc.
Gabon: Intergenerational Equity: Annuity Values (Permanent Income) Compared to FGF Scenarios
| 2005 | 2006 | 2010 | 2015 | 2020 | 2025 | 2030 | ||
|---|---|---|---|---|---|---|---|---|
| Constant total spending | ||||||||
| Scenario 1 (no FGF) 1/ | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | |
| Implied per capita spending 2/ | 76.8 | 75.3 | 69.5 | 63.0 | 57.0 | 51.7 | 46.8 | |
| Scenario 2 (FGF at 1.7 percent) 1/ | 85.4 | 85.4 | 85.4 | 85.4 | 85.4 | 85.4 | 85.4 | |
| Implied per capita spending 2/ | 65.7 | 64.4 | 59.5 | 53.9 | 48.8 | 44.2 | 40.0 | |
| Scenario 3 (FGF at 3.29 percent) 1/ | 101.1 | 101.1 | 101.1 | 101.1 | 101.1 | 101.1 | 101.1 | |
| Implied per capita spending | 77.8 | 76.2 | 70.4 | 63.8 | 57.8 | 52.3 | 47.4 | |
| Constant per capita spending | ||||||||
| Scenario 1 (no FGF) 2/ | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | |
| Implied total spending 1/ | 11.5 | 11.7 | 12.7 | 14.0 | 15.5 | 17.1 | 18.9 | |
| Scenario 2 (FGF at 1.7 percent) 2/ | 7.75 | 7.75 | 7.75 | 7.75 | 7.75 | 7.75 | 7.75 | |
| Implied total spending 1/ | 9.8 | 10.0 | 10.9 | 12.0 | 13.3 | 14.6 | 16.2 | |
| Scenario 3 (FGF at 3.29 percent) 2/ | 8.96 | 8.96 | 8.96 | 8.96 | 8.96 | 8.96 | 8.96 | |
| Implied total spending 1/ | 11.7 | 11.9 | 12.9 | 14.2 | 15.7 | 17.3 | 19.1 | |
Unit expressed in billion of 2005 CFA franc.
Unit expressed in thousand of 2005 CFA franc.
Alternative scenario: Contributions started in 1999
33. The scenarios discussed above assumed that contributions to the FGF are made from 2005 onwards, with a starting value of the FGF at CFAF 55.0 billion. However, the 1998 law called for contributions to begin 1999. Therefore, it is interesting to calculate how the outcome for current and future generations would have changed had the fund been built up since 1999. The analysis is repeated for the starting date of 1999; since the projected budget revenues for the years 1999-2004 are known, the actual windfall from oil revenues are used to calculate the contributions to the FGF. The new results are reported in Table 4. The corresponding results for the intergenerational equity calculation are found in Table 5.8
Gabon: Value of FGF, Total and Per Capita Consumption, With Contributions Starting in 1999
(Constant 2005 CFA franc)
Gabon: Value of FGF, Total and Per Capita Consumption, With Contributions Starting in 1999
(Constant 2005 CFA franc)
| Present Disc. Value (2005 CFAF) | 2005 | 2006 | 2010 | 2015 | 2020 | 2025 | 2030 | ||
|---|---|---|---|---|---|---|---|---|---|
| Oil Revenues (billion of CFA francs) | 4,401.3 | 731.7 | 617.6 | 315.7 | 156.9 | 77.0 | 41.9 | 23.2 | |
| Value FGF (billion of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | |
| Scenario 2 (FGF at 1.7 percent) | 1,010.1 | 1,090.7 | 1,333.9 | 1,832.0 | 2,079.5 | 2,134.6 | 2,125.3 | 2,094.6 | |
| Scenario 3 (FGF at 3.29 percent) | 2,263.9 | 1,109.4 | 1,356.9 | 1,879.2 | 2,164.7 | 2,260.5 | 2,291.9 | 2,301.1 | |
| Consumption (billion of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 4,401.3 | 731.7 | 617.6 | 315.7 | 156.9 | 77.0 | 41.9 | 23.2 | |
| Scenario 2 (FGF at 1.7 percent) | 3,966.1 | 431.2 | 381.9 | 241.7 | 142.4 | 89.5 | 60.7 | 45.0 | |
| Scenario 3 (FGF at 3.29 percent) | 5,219.9 | 440.9 | 395.3 | 263.3 | 168.5 | 117.5 | 89.7 | 74.7 | |
| Per capita consumption | |||||||||
| (Thousand of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 562.8 | 465.8 | 219.9 | 99.0 | 44.0 | 21.7 | 10.9 | ||
| Scenario 2 (FGF at 1.7 percent) | 331.7 | 288.0 | 168.4 | 89.9 | 51.1 | 31.4 | 21.1 | ||
| Scenario 3 (FGF at 3.29 percent) | 339.1 | 298.1 | 183.4 | 106.3 | 67.2 | 46.5 | 35.0 | ||
Gabon: Value of FGF, Total and Per Capita Consumption, With Contributions Starting in 1999
(Constant 2005 CFA franc)
| Present Disc. Value (2005 CFAF) | 2005 | 2006 | 2010 | 2015 | 2020 | 2025 | 2030 | ||
|---|---|---|---|---|---|---|---|---|---|
| Oil Revenues (billion of CFA francs) | 4,401.3 | 731.7 | 617.6 | 315.7 | 156.9 | 77.0 | 41.9 | 23.2 | |
| Value FGF (billion of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | |
| Scenario 2 (FGF at 1.7 percent) | 1,010.1 | 1,090.7 | 1,333.9 | 1,832.0 | 2,079.5 | 2,134.6 | 2,125.3 | 2,094.6 | |
| Scenario 3 (FGF at 3.29 percent) | 2,263.9 | 1,109.4 | 1,356.9 | 1,879.2 | 2,164.7 | 2,260.5 | 2,291.9 | 2,301.1 | |
| Consumption (billion of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 4,401.3 | 731.7 | 617.6 | 315.7 | 156.9 | 77.0 | 41.9 | 23.2 | |
| Scenario 2 (FGF at 1.7 percent) | 3,966.1 | 431.2 | 381.9 | 241.7 | 142.4 | 89.5 | 60.7 | 45.0 | |
| Scenario 3 (FGF at 3.29 percent) | 5,219.9 | 440.9 | 395.3 | 263.3 | 168.5 | 117.5 | 89.7 | 74.7 | |
| Per capita consumption | |||||||||
| (Thousand of CFA francs) | |||||||||
| Scenario 1 (no FGF) | 562.8 | 465.8 | 219.9 | 99.0 | 44.0 | 21.7 | 10.9 | ||
| Scenario 2 (FGF at 1.7 percent) | 331.7 | 288.0 | 168.4 | 89.9 | 51.1 | 31.4 | 21.1 | ||
| Scenario 3 (FGF at 3.29 percent) | 339.1 | 298.1 | 183.4 | 106.3 | 67.2 | 46.5 | 35.0 | ||
Gabon: Intergenerational Equity: Annuity Values (Permanent Income) Compared to FGF Scenarios, With Contributions Starting in 1999
Unit expressed in billion of 2005 CFA franc.
Unit expressed in thousand of 2005 CFA franc.
Gabon: Intergenerational Equity: Annuity Values (Permanent Income) Compared to FGF Scenarios, With Contributions Starting in 1999
| 2005 | 2006 | 2010 | 2015 | 2020 | 2025 | 2030 | ||
|---|---|---|---|---|---|---|---|---|
| Constant total spending | ||||||||
| Scenario 1 (no FGF) 1/ | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | |
| Implied per capita spending 2/ | 76.8 | 75.3 | 69.5 | 63.0 | 57.0 | 51.7 | 46.8 | |
| Scenario 2 (FGF at 1.7 percent) 1/ | 89.9 | 89.9 | 89.9 | 89.9 | 89.9 | 89.9 | 89.9 | |
| Implied per capita spending 2/ | 69.2 | 67.8 | 62.7 | 56.7 | 51.4 | 46.6 | 42.2 | |
| Scenario 3 (FGF at 3.29 percent) 1/ | 118.4 | 118.4 | 118.4 | 118.4 | 118.4 | 118.4 | 118.4 | |
| Implied per capita spending 2/ | 91.0 | 89.3 | 82.5 | 74.7 | 67.6 | 61.3 | 55.5 | |
| Constant per capita spending | ||||||||
| Scenario 1 (no FGF) 2/ | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | |
| Implied total spending 1/ | 11.5 | 11.7 | 12.7 | 14.0 | 15.5 | 17.1 | 18.9 | |
| Scenario 2 (FGF at 1.7 percent) 2/ | 7.97 | 7.97 | 7.97 | 7.97 | 7.97 | 7.97 | 7.97 | |
| Implied total spending 1/ | 10.4 | 10.6 | 11.4 | 12.6 | 14.0 | 15.4 | 17.0 | |
| Scenario 3 (FGF at 3.29 percent) 2/ | 10.5 | 10.5 | 10.5 | 10.5 | 10.5 | 10.5 | 10.5 | |
| Implied total spending 1/ | 13.6 | 13.9 | 15.1 | 16.6 | 18.4 | 20.3 | 22.4 | |
Unit expressed in billion of 2005 CFA franc.
Unit expressed in thousand of 2005 CFA franc.
Gabon: Intergenerational Equity: Annuity Values (Permanent Income) Compared to FGF Scenarios, With Contributions Starting in 1999
| 2005 | 2006 | 2010 | 2015 | 2020 | 2025 | 2030 | ||
|---|---|---|---|---|---|---|---|---|
| Constant total spending | ||||||||
| Scenario 1 (no FGF) 1/ | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | 99.8 | |
| Implied per capita spending 2/ | 76.8 | 75.3 | 69.5 | 63.0 | 57.0 | 51.7 | 46.8 | |
| Scenario 2 (FGF at 1.7 percent) 1/ | 89.9 | 89.9 | 89.9 | 89.9 | 89.9 | 89.9 | 89.9 | |
| Implied per capita spending 2/ | 69.2 | 67.8 | 62.7 | 56.7 | 51.4 | 46.6 | 42.2 | |
| Scenario 3 (FGF at 3.29 percent) 1/ | 118.4 | 118.4 | 118.4 | 118.4 | 118.4 | 118.4 | 118.4 | |
| Implied per capita spending 2/ | 91.0 | 89.3 | 82.5 | 74.7 | 67.6 | 61.3 | 55.5 | |
| Constant per capita spending | ||||||||
| Scenario 1 (no FGF) 2/ | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | 8.85 | |
| Implied total spending 1/ | 11.5 | 11.7 | 12.7 | 14.0 | 15.5 | 17.1 | 18.9 | |
| Scenario 2 (FGF at 1.7 percent) 2/ | 7.97 | 7.97 | 7.97 | 7.97 | 7.97 | 7.97 | 7.97 | |
| Implied total spending 1/ | 10.4 | 10.6 | 11.4 | 12.6 | 14.0 | 15.4 | 17.0 | |
| Scenario 3 (FGF at 3.29 percent) 2/ | 10.5 | 10.5 | 10.5 | 10.5 | 10.5 | 10.5 | 10.5 | |
| Implied total spending 1/ | 13.6 | 13.9 | 15.1 | 16.6 | 18.4 | 20.3 | 22.4 | |
Unit expressed in billion of 2005 CFA franc.
Unit expressed in thousand of 2005 CFA franc.
34. The main conclusions from the previous analysis continue to hold; that is: the FGF lifts consumption spending of future generations relative to a no-savings scenario, and future generations would stand to benefit more if the fund earned a market return.
35. A comparison of the numbers in Table 4 to those from Table 2 shows that the impact of an early start—1999 instead of 2005—is significant. For the low-remuneration scenario, the actualized value of the FGF is CFAF 1,010.1 billion compared with CFAF 556.2 billion, an increase of 81 percent. In the third scenario, the value of the FGF jumps from CFAF 1,247.8 billion to CFAF 2,263.9 billion. Second, the effects of increasing the consumption of future generations are stronger and kick in sooner. In both the second and third scenarios, per capita consumption under the early-savings case exceeds those from the late-savings case beginning as soon as 2010. Thus, if one accepts the premise that all oil revenues that should have been saved between 1999–2004 were in fact consumed and not used for investment spending or to pay down debt, then the results clearly illustrate how consumption early on comes at the expense of all future generations.
E. Policy Implications and Conclusions
36. The four main findings are summarized as follows. First, the Gabonese Fund for Future Generations is helpful in distributing consumption from present to future generations. Our scenarios indicate that most of the benefits would materialize from 2020 onwards.
37. Second, higher savings in the early years have a large payoff. The path of consumption and of the FGF were calculated with contributions starting in 2005 and were compared those to alternative scenarios in which contributions to the FGF started already in 1999. Significant differences among outcomes across the two starting dates were found, and the results underscore the importance of an early start. In Gabon, where oil production is projected to decline rapidly in the coming years, there is a sense of urgency for increasing savings to make up for lost time. A rapid build-up of net wealth will require substantial adjustment either in expenditures or in non-oil revenues, but doing so would ease the inevitable adjustment once oil revenues are near depletion.
38. Third, the rewards from saving for the future depend crucially on the institutional arrangements of the FGF, and on the rate of return on the assets in particular. A low return on the fund means that intergenerational redistribution is obtained at the cost of a lower potential consumption for all generations. It was shown that the value of the FGF under the current regime, in which it earns only 1.7 percent per annum, would improve dramatically if it earned a return that is closer to the one prevailing in international financial markets.
39. Fourth, the current institutional arrangement does not achieve intergenerational equity. To achieve such equity would require a substantial increase in savings in the short term, of a magnitude that may not be feasible in practice.
40. In the scenarios, the choice of helping future generations by paying down debt, instead of saving in a FGF, was not explicitly discussed. After all, either of the two choices would build up net financial wealth for future generations. It would be easy to include this part of the intertemporal dimension in our scenarios. For example, if it is assumed that most of the debt is contracted at the short-term interest rate, then the savings decision from the third scenario is equivalent to paying down debt. In fact, the average interest rate on the stock of external debt is 6.07 percent in 2005, clearly underscoring the relative attraction of using oil revenues primarily to pay down debt, at least in the short term. A moral case can be made that the present generation—to the extent that it contracted the debt and invested it poorly—should be primarily responsible for servicing it, i.e. that a larger non-oil primary surplus should be generated. But if this is not feasible, future generations would still be better off if oil revenues were used to reduce debt levels rather than accumulate deposits in the FGF.
Davis, Jeffrey, Rolando Ossowski, James Daniel, and Steven Barnett, 2001, “Stabilization and Savings Funds for Nonrenewable Resources. Experience and Policy Implications,” IMF Occasional Paper No 205 (Washington: International Monetary Fund),41 p.
Engel, Eduardo and Rodrigo Valdés, 2000, “Optimal Fiscal Strategies for Oil Exporting Countries,” IMF Working Paper 00/118 (Washington: International Monetary Fund), 70 p.
Fasano, Ugo, 2000, “Review of the Experience with Oil Stabilization and Savings Funds in Selected Countries,” IMF Working Paper 00/112 (Washington: International Monetary Fund), 21 p.
Katz, Menachem, Ulrich Bartsch, Harinder Malothra, and Milan Cuc, 2004, “Lifting the Oil Curse. Improving Revenue Management in Sub-Saharan Africa,” IMF, 65 p.
Rawls, John, 1971, A theory of Justice, Harvard University Press, 538 p.
Takizawa, Hajime, Edward Gardner, and Kenichi Ueda, 2004, “Are Developing Countries Better off Spending their Oil Wealth Upfront?” IMF Working Paper 04/141 (Washington: International Monetary Fund), 29 p.
Prepared by Chris Geiregat.
And the stock of productive physical assets. The issue if spending on investment goods is largely ignored in this paper to simplify the discussion.
For a discussion of savings to ensure justice across generations, see Rawls, 1971, pp. 251-258.
See Takizawa et al, 2004 for this argument. For theoretical models on fiscal strategies for oil producing countries, see Engel and Valdés, 2000.
The interested reader may find more information in Fasano, 2000, Davis et al., 2001; and Katz et al., 2004. This section draws heavily from those sources.
One way to understand this loss is as follows. Suppose that one could auction off all future oil revenues (plus the initial balance in the FGF) for one payment. An investor, using a discount rate of 3.29 percent, would offer CFAF 4,458.1 billion. However, if the contract required that investor to invest part of those future revenues in interest-bearing assets that yield only 1.7 percent, then she would offer only CFAF 3,766.4 billion. The difference is a net loss to the country.
Indeed, the per capita annuity is very sensitive to changes in the population growth rate. The per capita annuity will lie between one of two extremes. First, if g = r then no interest may be paid out, and second, if g = 0 then each period all permanent income is made available.
Of course, the first scenario yields the same numbers, as nothing is saved anyway. Thus we can limit our discussion to a comparison of the second and third scenarios.