I. Medium-Term Fiscal Sustainability in Trinidad and Tobago1
Trinidad and Tobago has greatly benefited from high energy prices in recent years. However, its reserves are being depleted rapidly and may be exhausted in 20 years at current extraction rates. Policy choices are complicated by uncertainties regarding future energy prices, real rates of return on financial assets, and the potential growth of the nonenergy sector. From a medium-term perspective, intergenerational distribution considerations call for accumulation of assets while smoothing the evolution of public expenditure is crucial from a macroeconomic management point of view. This paper discusses alternative ways of assessing medium-term fiscal sustainability and estimates the sustainable nonenergy deficit for Trinidad and Tobago using these different methodologies. The sensitivity of these estimates to shocks are also examined.
1. A dependence on revenues from nonrenewable natural resources poses a number of important challenges to fiscal policy. Fiscal policy needs to be geared towards ensuring that future generations can benefit from nonrenewable resource wealth notwithstanding the inevitable exhaustion of the resource endowment. This involves policy decisions on how much and at what pace energy wealth is transformed into other assets. However, expenditure decisions in that context need to take the economy’s absorption capacity into consideration, so as to minimize the risk of economic overheating and the erosion of wealth through inflation. At the same time, commodity prices tend to exhibit a high degree of volatility, thus implying that fiscal policy has a role to play in smoothing (or at least not exacerbating) economic cycles. Achieving these various objectives requires striking a delicate policy balance.
2. High energy prices, the discovery of new gas fields, and investment in downstream industries have significantly boosted energy revenues in Trinidad and Tobago. The increase in energy revenues, which have more than doubled relative to the level in the late 1990’s, has allowed for an impressive reduction in central government debt. However, energy revenue gains have also been used to increase social and capital expenditure and tax cuts to the nonenergy sector, resulting in a relatively significant deterioration in the underlying fiscal position.
|(In percent of GDP)|
3. Trinidad and Tobago’s own experience illustrates that the mismanagement of energy price booms can lead to severe economic hardship when prices decline. During the oil boom of the 1970s, the nonenergy deficit was allowed to soar from less than 10 percent of nonenergy GDP to over 40 percent by the early 1980s.2 When oil prices collapsed, the end of the oil boom brought about the need for a sharp fiscal adjustment and total expenditure had to be reduced from an average of 56 percent of nonenergy GDP in 1980–85 to 36 percent in the nineties. Output per capita declined by about 33 percent between 1981 and 1992, and unemployment rose from 10 percent to 22 percent in 1987–89.
4. This paper will focus on the challenge of distributing the nonrenewable resource wealth across generations. Its recommendations are geared towards the goal of intergenerational distribution and therefore focus on the transformation of the natural resource wealth into other assets. Issues related to the economy’s absorption capacity are beyond the scope of this paper and it is therefore assumed that the government accumulates financial assets as opposed to making other public infrastructure and social investments. However, the importance of taking cyclical and absorption considerations into account in the design of fiscal policy should not be underestimated.
B. Defining Fiscal Sustainability
5. Fiscal sustainability is a broader concept than financial solvency of the public sector. Fiscal solvency requires that the government’s intertemporal budget constraint is met, that is, that the net present value of future primary surpluses is greater or equal to the government’s initial current net debt. Fiscal sustainability requires that current revenue and expenditure policies can be maintained into the future while respecting the intertemporal budget constraint. This latter criterion safeguards against the need for drastic policy reversals—often associated with significant social and economic dislocation—when energy income declines. For countries whose stream of fiscal receipts is front loaded, as in the case of those endowed with nonrenewable resources, fiscal sustainability involves the accumulation of assets to finance public spending once the nonrenewable resource is exhausted.
6. A number of alternative criteria consistent with the above definition can be used to assess fiscal sustainability in countries with nonrenewable resources. Most criteria focus on a path for the nonenergy deficit, or for a path for expenditures given underlying assumptions for nonenergy revenues over time. The various criteria reflect different preferences for intergenerational distribution. The two most often cited criteria in the literature define fiscal policy as sustainable when the nonenergy deficit can be indefinitely maintained:
|Criterion I.||Constant in real terms. Under this criterion, energy wealth is converted into a fixed annuity stream in real terms to finance the nonenergy budget deficit indefinitely. Over time, as population and nonenergy output grow, the nonenergy deficit declines both in per capita terms and as percent of the nonenergy GDP. In practice, there is a need to reduce the nonenergy deficit as a share of nonenergy GDP.3|
|Criterion II.||Constant in percent of nonenergy GDP. Under this criterion, energy wealth is converted into a stream that finances a nonenergy deficit that can be maintained indefinitely as a percent of nonenergy GDP. The estimates in this case are subject to the additional uncertainty of having to project nonenergy GDP growth. In this more stringent sustainability criterion, preferences between the consumption of current and future generations are better balanced. Furthermore, contrary to Criterion I, there is no need for sustained fiscal adjustment to accommodate a sustainable nonenergy deficit that must shrink as a percent of GDP.4|
7. Estimating fiscal sustainability inevitably requires making assumptions with respect to a number of key parameters. The estimates presented in this paper are based on the following assumptions:
- Energy prices. Projections for energy prices reflect those underlying in the IMF’s May 2006 World Economic Outlook (WEO) until 2011. In particular, oil prices are assumed to decline gradually from US$66.5/bbl in 2006 to an average of US$66.0/bbl in 2011. Thereafter they converge towards a long-term value of US$45/bbl in real terms. This level is slightly above the historical average, reflecting the current understanding that some of the recent increases in oil prices is structural.5
- Energy reserves and extraction probabilities. Fiscal sustainability is highly sensitive on the overall endowment of energy reserves. For this exercise, data on energy reserves reflects the latest audited data on proven, probable, and possible reserves. However, proven, probable and possible reserves are assumed to have extraction probabilities of 100 percent, 75 percent and 50 percent, respectively, to reflect the uncertainty about the economic feasibility of extracting them. Figure 1 shows the baseline path of production and prices used in the exercise.
- Real interest rate. The real rate of return on assets is set at 4 percent, in line with the annual real return net of management costs for the Norwegian Petroleum Fund since 1997, of 4.3 percent.6 A more conservative investment policy tilted toward debt instruments would call for a lower real rate of return.
- Nonenergy GDP growth. The assumption for the growth rate of nonenergy GDP is 3 percent, in line with long-term growth rates of Caribbean countries.7
Figure 1.Trinidad and Tobago: Oil and Gas Production and Prices
Source: Fund Staff projections
|Real interest rate||4.0|
|Rate of growth of NE-GDP||3.0|
|Long-term oil price (US$/bbl)||45.0|
|Long-term gas price (US$/mcf)||5580.0|
|Oil reserves (Jan 05), in bbl mn|
|Gas reserves (Jan 04), in tcf|
8. Fiscal sustainability estimates are also based on policy parameters that describe the fiscal claim on energy and nonenergy sector GDP. While the government take from these sectors is subject to change, the working assumption for this paper is that their current values are a good indication of tax policy for Trinidad and Tobago over the medium term.
9. Presently, the government take from the energy sector is based on a percent share of the production value of crude oil and natural gas.8 The taxation regime for oil and gas are both based on production and income taxes. In periods of high energy prices, income taxes tend to yield more cents on the dollar of energy produced. Hence the government take is non-linear on energy prices. In FY 2005/06 there was a reform in the tax regime for oil and gas, with the goal of significantly increasing energy revenues. Preliminary figures for FY 2005/06 indicate that the government take has reached 27 percent of the production value of crude oil and natural gas—a substantial increase relative to about 18 percent in the previous 3 years. For the sake of simplicity and the purposes of this paper, the government take from the production value of crude oil and natural gas is assumed to remain at 27 percent indefinitely. A constant government take is likely to overestimate revenues when prices are low and underestimate it when prices are high. In that light, more prudence is called for when interpreting low-price scenarios.
|From nonenergy GDP||22.5||20.7|
|From energy sector||18.2||27.1|
10. The baseline value for the government take on the nonenergy sector is 20.7 percent of the nonenergy GDP. This figure is lower than the average for the fiscal years ending in 2001 and 2005, which was about 22½ percent of non-energy GDP, reflecting the FY2005/06 tax reform that included reductions in income and petrol excise taxes. Nevertheless, even at 20.7 percent of GDP, this exceeds the average tax revenue for other countries of similar income and populations which averages around 16 percent of GDP.
D. Fiscal Sustainability: Baseline
11. This section presents the nonenergy deficit, primary expenditure, and net asset profile that are sustainable under the two sustainability criteria presented above. It also discusses a scenario in which the nonenergy deficit is sustained at the level projected for the current fiscal year and then brought abruptly into balance in 2020 is also examined.
Criterion I: Nonenergy Deficit Constant in Real Terms
12. The sustainable nonenergy deficit in the baseline is TT$ 11,724 million in FY 2004/05 prices. In FY 2005/06 the sustainable nonenergy deficit is 18¼ percent of current nonenergy GDP, or 10¾ of current GDP. This level is lower than the projected primary nonenergy deficit based on the supplemental budget of FY 2005/06 (23½ percent of nonenergy GDP, or 13½ percent of GDP). Sustainability in this case entails a building-up net assets equivalent to 135 percent of nonenergy GDP by 2044, when oil and gas reserves are exhausted in the baseline scenario. After that, the real value of the fund will be constant in real terms and withdrawals from the fund will equal the real return earned by the fund net assets.
13. Assuming that the government take from the nonenergy sector is constant and equal to 21 percent of nonenergy GDP, the ratio of primary public expenditure to nonenergy GDP will converge in the long-run to the government take since the transfer from the revenue fund will shrink relative to the growing size of the nonenergy economy.
Case 1. Constant non-energy deficit in real terms
14. In the 2005 Article IV Consultation (IMF 2005), the IMF estimated the sustainable nonenergy deficit in constant dollar terms for Trinidad and Tobago at about 13½ percent of FY 2005/06 GDP—substantially higher than what is estimated in this paper. The difference with the 10¾ percent estimate in this paper reflects: (i) the introduction of varying extraction probabilities of energy reserves; (ii) a lower government take on energy output over the projection period in line with recent history; (iii) upward revisions to GDP; and (iv) a more conservative medium term energy prices. The impact of these changes was only partially offset by higher energy prices until 2011 (as projected in the WEO) and a higher assumed real rate of return.
Criterion II: Nonenergy deficit constant in percent of nonenergy GDP
15. Since the nonenergy sector will likely continue to grow at positive rates in the long-run, the sustainable nonenergy deficit in percent of nonenergy GDP is substantially lower than the sustainable deficit in constant dollar terms estimated using criterion I. Under criterion II, the sustainable nonenergy deficit is about 4¼ percent of the nonenergy GDP. For FY 2005/06, this entails a sustainable nonenergy deficit of TT$2,850 million in current prices, which is substantially lower than the nonenergy deficit currently projected based on the FY2005/06 supplemental budget.
16. This sustainable path would entail an accumulation of net assets in a nonrenewable resource fund building-up to 430 percent of nonenergy GDP.9 Assuming a constant government take from nonenergy GDP equal to 21 percent, primary public expenditure is projected to stay constant at 25¼ percent of nonenergy GDP indefinitely.
Case 2. Constant non-energy deficit as percent of non-energy GDP
Zero Deficit in 2020
17. Maintaining the nonenergy deficit at the level currently projected for FY 2005/06 involves a drastic fiscal adjustment. Balancing the budget in 2020 when energy revenues are near exhaustion and accumulated assets are not enough to finance any further significant deficits would require an adjustment of about 30 percent of nonenergy GDP.
18. In this case, net assets in the nonrenewable resource fund would peak at 26 percent of nonenergy GDP in 2012 and then would drop to zero by 2020, when the nonenergy deficit would converge to the energy revenue. This strategy would entail a higher level of primary expenditure than the policy of sustainable nonenergy deficit constant in dollar terms until 2021, and lower thereafter. How to attain the necessary fiscal adjustment to achieve zero deficit in 2020 is beyond the scope of this paper, and it is not clear that the fiscal adjustment necessary to balance the budget in 2020 can be attained without a hard landing.
Case 3. Zero overall deficit in 2020
E. Sensitivity Analysis and Stress Tests
19. Fiscal sustainability also requires that fiscal targets are robust to most changes in assumptions. Changes in key macroeconomic parameters underlying directly affect estimates of the sustainable nonenergy deficit. However, fiscal targets are generally deemed robust when they can accommodate likely deviations from baseline parameters. Table 3 below presents low and high parameter values for sensitivity analysis. In particular: (i) the long-term crude oil price, which is assumed to be constant at US$45/bbl beyond 2011, is set at US$25 for the most pessimistic and US$65 in the most optimistic price scenarios; (ii) the return on the net assets of the revenue fund is set between 3.5 percent and 4.5 percent; (iii) the growth rate of nonenergy GDP is set between 2.5 percent and 3.5 percent; and, (iv) the stock of crude oil and natural gas reserves is bounded between a low reserves scenario in which only proven reserves are available and a high reserves one in which the full stock of proven, possible and probable reserves is available.
|Case 1||Case 2|
|Description||FY 2005/06||FY 2025/26||FY 2065/66||Indefinitely|
|Pessimistic oil prices||US$25 long-run oil price||14.4||7.3||2.2||3.3|
|Low oil prices||US$35 long-run oil price||16.3||8.3||2.5||3.7|
|High oil prices||US$55 long-run oil price||20.0||10.2||3.1||4.6|
|Optimistic oil prices||US$65 long-run oil price||21.9||11.1||3.4||5.0|
|Low rate of return||3.5% real rate of return||16.6||8.4||2.6||2.2|
|High rate of return||4.5% real rate of return||19.5||9.9||3.0||6.0|
|Low nonenergy GDP growth||2.5% growth in long-run||18.2||9.9||3.7||6.1|
|High nonenergy GDP growth||3.5% growth in long-run||18.2||8.6||2.2||2.1|
|Low reserves||Only proven reserves||12.7||6.4||2.0||2.9|
|High reserves||100% of proven, possible and probable||20.5||10.4||3.2||4.7|
20. Fiscal sustainability estimates are generally robust to changes in individual assumptions. Table 3 shows the sustainable nonenergy deficit for FY 2005/06, 2025/26 and 2065/66 obtained when a single parameter is varied while keeping other parameters constant, for the Criterion I sustainable deficit in constant dollar terms (Case 1), as well as the Criterion II sustainable deficit as constant share of nonenergy GDP (Case 2). The main results are as follows:
- Changes in the crude oil price assumption. When the assumption for long-term crude oil prices are changed between US$35 per barrel and US$65 per barrel, the sustainable nonenergy deficit varies from 14.4 percent to 21.9 percent of current nonenergy GDP in Case 1, and from 3.3 percent to 5.0 percent of nonenergy GDP in Case 2.
- Changes in the real interest rate assumption. The real interest rate determines the return on the net assets accumulated in the revenue fund. As real interest rates range from 3½ percent to 4½ percent, the sustainable nonenergy deficit goes from 16.6 percent to 21.9 percent of current nonenergy GDP in Case 1, and 2.2 percent to 6.0 percent of nonenergy GDP in Case 2.
- Changes in the nonenergy GDP growth assumption. The nonenergy GDP growth assumption does not affect the sustainable nonenergy deficit in real terms in Case 1, but it affects its size relative to the nonenergy GDP in the future. Because the faster the nonenergy GDP grows the greater the need for asset accumulation to finance a growing nonenergy deficit, the growth rate of the nonenergy GDP is an important determinant of the sustainable nonenergy deficit in Case 2. As the nonenergy GDP growth assumption increases from 2½ percent to 3½ percent, the sustainable nonenergy deficit from 6.1 percent to 2.1 percent of nonenergy GDP.
- Changes in the oil and gas reserves assumption. In a low reserves scenario, probable and possible reserves are assumed to have a zero extraction probability; in a high reserves scenario, all reserves receive a 100 percent extraction probability. In moving from the low to the high reserves scenario, the sustainable nonenergy deficit goes from 12.7 percent to 20.5 percent of FY 2005/06 current nonenergy GDP in Case 1, and from 2.9 percent to 4.7 percent of nonenergy GDP in Case 2.
21. The possibility that shocks occur to more than one parameter can also be assessed. In the worst case scenario, the real interest rate, reserves are concomitantly low, and energy prices are pessimistic (as defined in Table 3). Conversely, in the best case scenario, the real interest rate, reserves are concurrently high, and energy prices are optimistic. Table 4 presents the sustainable nonenergy deficit for FY 2005/06 for the best and worst case scenarios. In the worst case scenario, the sustainable deficit is 10.8 percent of nonenergy GDP for Criterion I and 1.4 percent of nonenergy GDP for Criterion II.
|Worst Case||Best Case|
22. Based on the above analysis, fiscal sustainability in Trinidad and Tobago is at risk. Preliminary estimates suggest that the projected outturn for FY 2005/06 is sustainable only under the best case scenario for the parameters when sustainability is defined as a constant nonenergy deficit in real terms—the weaker of the two sustainability criteria examined. In the best case scenario, the sustainable nonenergy deficit reaches 24.4 percent of current nonenergy GDP (Criterion I), which is slightly larger than the 23½ percent currently being projected based on the supplemental budget for FY 2005/06. However, under the most stringent rule which requires a constant nonenergy deficit as percent of nonenergy GDP (Criterion 2), the sustainable nonenergy deficit is 7½ which is significantly smaller than the projected figures for FY 2005/06.
23. Intergenerational distribution requires that the consumption of nonrenewable wealth be spread over time and across generations. This paper presented two criteria for determining fiscal sustainability in the presence of revenues stemming from a nonrenewable resource.
24. Estimates for Trinidad and Tobago suggest that the projected nonenergy deficit in FY 2005/06 is likely to exceed the sustainable level under both criteria. For the least stringent of the two criteria, which requires the nonenergy deficit to be constant in real terms, current fiscal performance is only sustainable under a best case scenario with high medium-term energy prices, reserves, and return on assets.
BailenG. and V.Kramarenko2004“Issues in Medium-Term Management of Oil Wealth,”IMF Country Report No. 04/308 (Washington: International Monetary Fund) pp. 22–36.
BarnettSteven and RolandoOssowski2003“Operational Aspects of Fiscal Policy in Oil-Producing Countries” in Fiscal Policy Formulation and Implementation in Oil-Producing Countriesed. byJeffreyDavisRolandoOssowski and AnnaliseFedelino (Washington: International Monetary Fund) pp. 45–81.
BarnettSteven and AlvaroVivanco2003“Statistical Properties of Oil Prices: Implications for Calculating Government Wealth” in Fiscal Policy Formulation and Implementation in Oil-Producing Countriesed. byJeffreyDavisRolandoOssowski and AnnaliseFedelino (Washington: International Monetary Fund) pp. 123–149.
HusainAasim and HamidDavoodi2005“How Should Middle Eastern and Central Asian Oil Exporters Use Their Oil Revenues?” in International Monetary Fund World Economic Outlook: Chapter I: Economic Prospects and PoliciesApril2005 (Washington: International Monetary Fund) pp.54–56.
International Monetary Fund2002“Assessing Sustainability” mimeo.
International Monetary Fund2005“Long-Term Sustainability”in IMF Country Report No. 06/32 (Washington: International Monetary Fund) p. 12.
JafarovEtibar and KenjiMoriyama2005“The Norwegian Government Petroleum Fund and the Dutch Disease,”IMF Country Report No. 05/197 (Washington: International Monetary Fund) pp. 39–55.
KimYitae Kevin2005“Managing Oil/Gas Wealth in Timor-Leste,”IMF Country Report No. 05/250 (Washington: International Monetary Fund) pp. 16–29.
LeighDaniel and Jan-PeterOlters2006. “Natural Resource Depletion, Habit Formation, and Sustainable Fiscal Policy: Lessons from Gabon,”IMF Country Report No. 06/232 (Washington: International Monetary Fund) pp. 4–21.
LeiteCarlos2004. “Prudent Management of Oil Wealth,”IMF Country Report No. 04/231 (Washington: International Monetary Fund) pp. 41–45.
LohmusPeter2005. “Fiscal Management of Kazakhstan’s Oil Wealth,”IMF Country Report No. 05/240 (Washington: International Monetary Fund) pp. 12–28.
MishraPrachi2006“Emigration and Brain Drain: Evidence from the Caribbean,”IMF Working Paper 06/25 (Washington: International Monetary Fund).
Norges Bank2005“Management of the Government Petroleum Fund: Report for the Third Quarter of 2005” (Oslo: Norges Bank).
SeguraAlonso2006. “Management of Oil Wealth Under the Permanent Income Hypothesis: The Case of São Tomé and Príncipe,”IMF Working Paper 06/183 (Washington: International Monetary Fund).
TakizawaHajime2005. “Fiscal Sustainability and Options for Fiscal Adjustment,”IMF Country Report No. 05/234 (Washington: International Monetary Fund) pp. 25–39.
TakizawaHajimeEdward H.Gardner and KenichiUeda2004. “Are Developing Countries Better Off Spending Their Oil Wealth Upfront?,”IMF Working Paper 04/141 (Washington: International Monetary Fund).
VelculescuDelia and SaqibRizavi2005. “Trinidad and Tobago: The Energy Boom and Proposals for a Sustainable Fiscal Policy,”IMF Working Paper 05/197 (Washington: International Monetary Fund).