## _wp1028 ## Source details **Canonical URL:** [_wp1028](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp1028.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp1028.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp1028.pdf.json) --- ### Summary of methodological points and indicators - Key indicators emphasized for OPCs: - Overall balance (OB): useful for net financing requirement but can mask fiscal expansion financed by oil revenue. - Non-oil balance: excludes net oil revenue and better captures fiscal impact on domestic demand and the injection/use of oil revenue. - Non-oil primary balance (NOPB): further excludes interest receipts/payments and is informative for fiscal sustainability comparisons. - Cyclically-adjusted non-oil primary balance: excludes automatic stabilizers to isolate discretionary fiscal policy. - Ratios should be scaled to non-oil GDP in OPCs to avoid distortions from oil-price driven nominal GDP movements. ### Main findings on fiscal outcomes during 2003–2008 (the “boom”) - Overall fiscal balances: - The simple average overall surplus for the whole sample rose from 2 percent of GDP in 2003 to 12 percent of GDP in 2008. - The average overall balance improved by 2½ percent of GDP in 2008, when oil prices jumped to close to US$100 per barrel. - About a dozen countries recorded a deterioration in their overall fiscal balances despite the broad average improvement; three countries (Ecuador, Iran, and Yemen) shifted from fiscal surpluses in 2007 to deficits in 2008. - Use versus saving of additional oil revenue: - Between 2004 and 2008, the cumulative change in the non-oil primary deficit in nominal terms represented, on average, 54 percent of the cumulative change in oil (relative to the 2003 level). - Utilization rates by income level (2004–08, in percent of additional oil revenue): - Low income: 78 percent (use) - High income: 45 percent (use) - On average, OPCs “used” about half of the increased oil revenue during the period. - Non-oil primary balance and primary spending: - The simple average non-oil primary deficit, which had fluctuated around 27 percent of non-oil GDP since the mid-1990s, increased sharply to about 40 percent of non-oil GDP by 2008. - Primary spending rose by an average of 14 percentage points of non-oil GDP between 2003 and 2008 (from 44 to 58 percent of non-oil GDP). - Excluding Libya and Timor-Leste, the increase in primary spending was 9 percentage points of non-oil GDP. - Investment (capital expenditure) grew by 9 percentage points of non-oil GDP for the whole sample; excluding Libya and Timor-Leste, the increase was 5 percentage points of non-oil GDP. - Real growth rates of spending (three-year moving averages): - Total government spending growth increased from 9 percent in 2003 to 17 percent in 2008. - Capital expenditure growth accelerated from 15 percent to 28 percent over the same period. - Composition by income group (changes 2003–08, in percent of non-oil GDP): - Primary spending: - High: 7.4% - Upper-middle: 5.9% - Lower-middle: 11.9% - Low: 7.1% - o/w capital spending: - High: 6.3% - Upper-middle: 1.7% - Lower-middle: 8.6% - Low: -1.0% - o/w current primary spending: - High: 1.2% - Upper-middle: 4.1% - Lower-middle: 3.3% - Low: 8.2% - Cross-group patterns: - Fiscal expansions were primarily driven by increases in primary spending, particularly capital spending in high- and lower-middle-income OPCs. - Increased current spending was the key driver in low-income OPCs (and to some extent in upper-middle income OPCs), with capital spending ratios largely flat for low-income countries. - Explanatory factors for high “usage” of additional oil revenue in low-income OPCs include revenue earmarking and coparticipation provisions (e.g., Nigeria) and increases in explicit fuel subsidies (e.g., Yemen). ### Methodological illustration: Nigeria (2003–09) - Nigeria fiscal indicators illustrate methodological issues: - Improvement in the overall balance in 2004–05 hid that spending financed by oil revenue increased (as shown by deteriorating NOPB). - The NOPB ratio to total GDP can fail to show the extent of expansion in 2004–05 and contraction in 2009 because total GDP (the denominator) moves with oil prices. ### Implications from the boom period - Standard fiscal indicators must be refined (use of non-oil and cyclically-adjusted measures and scaling by non-oil GDP) to assess fiscal stance accurately in OPCs. - The boom period (2003–08) featured large fiscal expansions in OPCs, financed largely by higher oil revenues, with substantial variation across income groups in composition and degree of revenue “use.” - Low-income OPCs tended to convert a larger share of additional oil revenue into increased spending, particularly current spending. ### 2009: oil price shock and fiscal outcomes - Early-2009 oil price dynamics and immediate effects: - Oil prices hovered around US$40 per barrel for a couple of months early in the year, which led to a lot of concerns in finance ministries throughout OPCs, and liquidity problems in some of those countries. - Overall fiscal balances (projected for 2009): - The average overall surplus for the whole sample of countries would fall 10 percentage points of GDP to 2 percent of GDP in 2009, getting back to the level recorded in 2003. - The fiscal position is projected to shift from surplus to deficit in twelve countries in the sample. - The deterioration is more or less the same across country groups by income level, but more dramatic for high income countries, including GCC countries (14 percentage points of GDP). - Non-oil primary balance and expenditure composition (2008–09 changes): - Non-oil primary balance (percent of non-oil GDP) is projected to improve in lower income countries. - High income countries on average would expand their non-oil deficits. - Lower-middle income and low income countries would contract their deficits by an average of 4½ percent of non-oil GDP. - Most of the fiscal adjustment (about 4 percent of non-oil GDP) would come from a reduction in current expenditure. - Current expenditure would fall by 8 percent in real terms in 2009. - Part of the reduction in current expenditure would be endogenous: reductions in explicit fuel subsidies given lower international fuel prices (examples: Bolivia, Iran, Yemen) and reduced oil revenue coparticipation across various levels of government (examples: Nigeria, Sudan, Venezuela). - The reduction in revenue coparticipation implies “real” cuts in spending by subnational governments. - Capital expenditure in lower income countries would rise on average by 2 percent in real terms. - Table 3. OPCs: Expenditure ratios by income groups (changes 2008-09, in percent of non-oil GDP) - High: Primary spending 1.5% - o/w capital spending 3.3% - o/w current primary spending -1.8% - Upper-middle: Primary spending -1.4% - o/w capital spending -1.1% - o/w current primary spending -0.3% - Lower-middle: Primary spending -4.1% - o/w capital spending -0.6% - o/w current primary spending -3.5% - Low: Primary spending -7.0% - o/w capital spending -0.6% - o/w current primary spending -6.4% ### Prospects for 2010 based on futures-implied oil prices - IMF WEO projection: - IMF WEO is projecting a recovery in the average annual oil price for 2010 to US$77 per barrel based on observed future prices. - Preliminary 2010 projections: - Recovery of the overall fiscal balance by 4 percent of GDP on average. - Improvement more marked for high and low income OPCs. - Most of the improvement would come from higher oil revenue, but non-oil revenue ratios would also recover slightly in higher income countries together with global economic activity. - Primary spending ratios to non-oil GDP would, on average, fall slightly relative to 2009 levels. - Assumes some reshuffling within current expenditure items as intergovernmental transfers and fuel subsidies would rise together with oil prices. - Note on Mexico’s 2009 experience: - Mexico hedged oil export revenue in 2009 at US$70/barrel through a put option; this operation cost the government US$1.5 billion. - Its hedging program has been in place since 1991. - There is no counterparty risk as this operation is collateralized by US Treasuries. ### Fiscal policy in OPCs over the economic cycle: definitions, measurement, and cyclicality - Context and definitions: - Fiscal policy in OPCs has often been defined as expansionary/contractionary when the change in the non-oil primary balance (NOPB) is negative/positive. - Decomposition: ∆NOPB = ∆CNOPB + ∆CANOPB - CNOPB: cyclical non-oil primary balance - CANOPB: cyclically adjusted (structural) non-oil primary balance - Fiscal impulse (FI) = change in the cyclically adjusted non-oil primary deficit (∆CANOPB). - Automatic stabilizer (AS) = change in the cyclical non-oil primary deficit (∆CNOPB). - Authors redefine expansionary/contractionary based on the sign of ∆CANOPB (policy actions under policymakers’ control). - Cyclicality classification: - Procyclical: expansionary/contractionary fiscal policy when the change in the output gap is positive/negative. - Countercyclical: expansionary/contractionary fiscal policy when the change in the output gap is negative/positive. - Output gap measurement and estimation: - Output gap (OG) = (Y – Yp) / Yp, where Y is actual non-oil real GDP and Yp is potential non-oil real GDP. - Potential non-oil real GDP is estimated using an HP filter with smoothing parameter lambda = 100 for all countries; IMF WEO output projections used until 2014 to reduce end-point sensitivity. - Fiscal impulses estimated following Horton, Kumar and Mauro (2009) and Fedelino, Ivanova, and Horton (2009) as changes in the cyclically-adjusted non-oil primary deficit measured in relation to potential (non-oil) output. - Elasticity assumptions: elasticities of non-oil revenue and non-oil primary expenditure with respect to the output gap are one and zero, respectively. ### Empirical findings on cyclicality - 2003–2008 (boom): - Fiscal policy was generally procyclical during the boom period (2003-2008). - Change in the output gap was positive in all subgroups and was more pronounced in upper middle income (UMI) countries. - Fiscal impulses were positive in all subgroups during 2003-2008 and especially large in UMI countries. - Combination of positive changes in output gaps with positive fiscal impulses implies a procyclical fiscal policy response. - Degree of procyclicality (FI/ΔOG) is higher for low income (LI) countries and lower for high income (HI) countries. - Reverse causality caveat: literature typically assumes output shocks drive fiscal policy; some authors claim fiscal policy shocks drive output. Evidence of procyclical fiscal policy in developing countries is robust to endogeneity considerations. - 2009 downturn: - Change in the output gap was negative in all subgroups in 2009 and was more substantial in UMI countries. - Fiscal impulses were negative in all subgroups except in HI countries, where on average fiscal policy was neutral. - Negative changes in output gaps coupled with negative fiscal impulses imply a procyclical fiscal policy response in 2009. - Degree of procyclicality remained higher for LI countries. - Comparative pattern: procyclical response of LI countries during 2009 was stronger than during 2003-2008; for other subgroups the procyclical response was stronger during 2003-2008 than in 2009. - Preliminary projections for 2010 suggest a procyclical fiscal policy stance for all country groups, with some fiscal tightening accompanying negative changes in output gaps; all country groups would display relatively similar degrees of procyclicality in 2010. - Quantitative measure of procyclicality (selected reported figures from Table 4): - HI: ΔOG 7.0; FI 7.8; FI/ΔOG 1.1 (2003-2008) and ΔOG -3.3; FI 0.0; FI/ΔOG 0.0 (2009) - UM: ΔOG 10.3; FI 13.6; FI/ΔOG 1.3 (2003-2008) and ΔOG -7.1; FI -4.6; FI/ΔOG 0.7 (2009) - LM: ΔOG 3.8; FI 11.7; FI/ΔOG 3.1 (2003-2008) and ΔOG -2.1; FI -4.5; FI/ΔOG 2.1 (2009) - L: ΔOG 2.6; FI 9.0; FI/ΔOG 3.4 (2003-2008) and ΔOG -1.6; FI -8.0; FI/ΔOG 5.1 (2009) ### Vulnerability to declines in oil prices: approach and key results - Approach and caveats: - Simplified linear approach: linear adjustment of oil revenue and oil GDP in the 2009 projections, keeping non-oil revenues and total spending ratios to non-oil GDP unchanged. - Caveats: assumes local linearity between oil prices and fiscal oil revenue and abstracts from policy responses (exchange rate movements, non-oil revenue increases, spending cuts, changes in intergovernmental transfers, reductions in fuel subsidies). - Fiscal impact estimates for a US$10 per barrel shock: - The overall fiscal balance in a subsample of OPCs would fall on average by 3½ percent of GDP in response to a US$10 per barrel lower price than in the baseline. - Starting from 2009 projected figures, 18 out of the 21 countries in the analyzed subsample would record overall fiscal deficits after the shock. The average deficit would increase to 7 percent of GDP. - This average deficit would be larger than the one recorded in 1998, when oil prices in real terms recorded their lowest level in the period since 1973 (US$17 per barrel at 2009 prices). - Financing and reserves implications: - On average, the 18 OPCs in the sample recording overall fiscal deficits would have to use about half of their stock of international reserves plus foreign assets held in oil funds projected to be available at the end of 2008. - Import coverage of international reserves would fall below 3 months of imports in 5 countries. - If there was no fiscal adjustment and/or a depreciation of the currency and oil prices remained low for an additional year, reserve coverage would fall below the 3 month import coverage threshold in half the countries in the subsample. - Countries considered for the US$10 shock exercise: - Algeria, Angola, Azerbaijan, Bolivia, Cameroon, Chad, Congo, Ecuador, Equatorial Guinea, Gabon, Indonesia, Iran, Kazakhstan, Mexico, Nigeria, Russia, Sudan, Trinidad and Tobago, Venezuela, Vietnam, and Yemen. - Countries with quite substantial foreign assets were excluded from this analysis. ### Fiscal sustainability: methodology and findings - Key sustainability concept: - Incorporate exhaustibility of oil reserves and government net financial position. - Ten countries had ratio of total proved reserves to (current) production of less than 25 years by end-2008 (based on British Petroleum proved reserves). - Principal uncertainties in OPCs: oil reserves, oil prices and volumes, costs of production, government take, sale of gas reserves contingent on contracts and infrastructure, and statistical properties of the oil price. - Sustainability benchmark methodology: - Step 1: compute government net wealth as present value of projected future oil revenues (evaluated at oil prices prevailing in the respective year of analysis) plus net government financial assets. - Step 2: derive a consumption/spending path (the NOPD) as an annuity over the remaining production period (reserves-to-production ratio in years) plus 15 years. - Discounting and parameters: annual proved oil reserves from BP; constant real oil prices at year-of-analysis level; 4-year average government take from oil production; interest rate of 3 percent in real terms to discount future oil revenue flows and to calculate annuities. - Rationale: annuity over remaining production period plus 15 years differs from the usual perpetuity benchmark and is chosen to be more appropriate for countries with short oil production horizons. - Interpretation and main empirical finding: - For countries with short production horizons (lower than 25 years), the methodology implies existing government wealth would be consumed over 1 generation (35-40 years), including 15 years to adjust for oil exhaustion, implying a sharp fiscal adjustment immediately after the end of such generation. - Most OPCs recorded a deterioration in their sustainability position between 2003 and 2009, mainly because of a larger expansion in their NOPD relative to increases in government net wealth. - Summary indicator: “fiscal sustainability ratio” computed as the ratio of implied sustainable or long-term primary expenditure relative to actual primary expenditure. Countries recording a fiscal sustainability ratio lower than 1 would have to adjust to reach the sustainable benchmark. - Cross-income variation: - Lower income countries, on average, were short of their sustainability benchmarks in both 2003 and 2009; the position deteriorated in low income countries between 2003 and 2009. - Lower-middle income countries improved between 2003 and 2009 (improvement driven by the Republic of Congo; excluding Congo, lower-middle income countries would have shown an unchanged fiscal sustainability position). - Higher income countries’ fiscal stances were in line with the implied sustainability benchmark in 2009 (just at the horizontal axis), but deteriorated between 2003 and 2009. *Italic source: _wp1028 - IMF working paper content provided in the supplied PDF excerpt.* ### 1. Expenditure Ratios by Income Groups (changes 2003–2008 in percent of non-oil GDP)..10 ### 1. Expenditure Ratios by Income Groups (changes 2003–2008 in percent of non-oil GDP) ### Summary of methodological points and indicators - Key indicators emphasized for OPCs (oil producing countries): - Overall balance (OB): useful for net financing requirement but can mask fiscal expansion financed by oil revenue. - Non-oil balance: excludes net oil revenue and better captures fiscal impact on domestic demand and the injection/use of oil revenue. - Non-oil primary balance (NOPB): further excludes interest receipts/payments and is informative for fiscal sustainability comparisons. - Cyclically-adjusted non-oil primary balance: excludes automatic stabilizers to isolate discretionary fiscal policy. - Ratios should be scaled to non-oil GDP in OPCs to avoid distortions from oil-price driven nominal GDP movements. ### Main findings on fiscal outcomes during 2003–2008 (the “boom”) - Overall fiscal balances: - The simple average overall surplus for the whole sample rose from 2 percent of GDP in 2003 to 12 percent of GDP in 2008. - The average overall balance improved by 2½ percent of GDP in 2008, when oil prices jumped to close to US$100 per barrel. - About a dozen countries recorded a deterioration in their overall fiscal balances despite the broad average improvement; three countries (Ecuador, Iran, and Yemen) shifted from fiscal surpluses in 2007 to deficits in 2008. - Use versus saving of additional oil revenue: - Between 2004 and 2008, the cumulative change in the non-oil primary deficit in nominal terms represented, on average, 54 percent of the cumulative change in oil (relative to the 2003 level). - Utilization rates by income level (2004–08, in percent of additional oil revenue): - Low income: 78 percent (use) - High income: 45 percent (use) - On average, OPCs “used” about half of the increased oil revenue during the period. - Non-oil primary balance and primary spending: - The simple average non-oil primary deficit, which had fluctuated around 27 percent of non-oil GDP since the mid-1990s, increased sharply to about 40 percent of non-oil GDP by 2008. - Primary spending rose by an average of 14 percentage points of non-oil GDP between 2003 and 2008 (from 44 to 58 percent of non-oil GDP). - Excluding Libya and Timor-Leste, the increase in primary spending was 9 percentage points of non-oil GDP. - Investment (capital expenditure) grew by 9 percentage points of non-oil GDP for the whole sample; excluding Libya and Timor-Leste, the increase was 5 percentage points of non-oil GDP. - Real growth rates of spending (three-year moving averages): - Total government spending growth increased from 9 percent in 2003 to 17 percent in 2008. - Capital expenditure growth accelerated from 15 percent to 28 percent over the same period. - Composition by income group (changes 2003–08, in percent of non-oil GDP; Table 2): - Primary spending: - High: 7.4% - Upper-middle: 5.9% - Lower-middle: 11.9% - Low: 7.1% - o/w capital spending: - High: 6.3% - Upper-middle: 1.7% - Lower-middle: 8.6% - Low: -1.0% - o/w current primary spending: - High: 1.2% - Upper-middle: 4.1% - Lower-middle: 3.3% - Low: 8.2% - Cross-group patterns: - Fiscal expansions were primarily driven by increases in primary spending, particularly capital spending in high- and lower-middle-income OPCs. - Increased current spending was the key driver in low-income OPCs (and to some extent in upper-middle income OPCs), with capital spending ratios largely flat for low-income countries. - Explanatory factors for high “usage” of additional oil revenue in low-income OPCs include revenue earmarking and coparticipation provisions (e.g., Nigeria) and increases in explicit fuel subsidies (e.g., Yemen). ### Representative country example (methodological illustration) - Nigeria fiscal indicators (2003–09) illustrate methodological issues: - Improvement in the overall balance in 2004–05 hid that spending financed by oil revenue increased (as shown by deteriorating NOPB). - The NOPB ratio to total GDP can fail to show the extent of expansion in 2004–05 and contraction in 2009 because total GDP (the denominator) moves with oil prices. ### Implications highlighted in the text - Standard fiscal indicators must be refined (use of non-oil and cyclically-adjusted measures and scaling by non-oil GDP) to assess fiscal stance accurately in OPCs. - The boom period (2003–08) featured large fiscal expansions in OPCs, financed largely by higher oil revenues, with substantial variation across income groups in composition and degree of revenue “use.” - Low-income OPCs tended to convert a larger share of additional oil revenue into increased spending, particularly current spending. *Italic source: _wp1028 - 1. Expenditure Ratios by Income Groups (changes 2003–2008 in percent of non-oil GDP)..10* ### 2009. Oil prices hovered around US$40 per barrel for a couple of months early in the year, ### _wp1028 - 2009. Oil prices hovered around US$40 per barrel for a couple of months early in the year, ### 2009: oil price shock and fiscal outcomes - Oil prices hovered around US$40 per barrel for a couple of months early in the year, which led to a lot of concerns in finance ministries throughout OPCs, and liquidity problems in some of those countries. - Overall fiscal balances (measured in percent of GDP) are projected to worsen substantially in 2009. - The average overall surplus for the whole sample of countries would fall 10 percentage points of GDP to 2 percent of GDP in 2009, getting back to the level recorded in 2003. - The fiscal position is projected to shift from surplus to deficit in twelve countries in the sample. - The deterioration is more or less the same across country groups by income level, but more dramatic for high income countries, including GCC countries (14 percentage points of GDP). ### Non-oil primary balance and expenditure composition (2008–09 changes) - In contrast, the non-oil primary balance (measured in percent of non-oil GDP) is projected to improve in lower income countries. - High income countries on average would expand their non-oil deficits. - Lower-middle income and low income countries would contract their deficits by an average of 4½ percent of non-oil GDP. - Most of the fiscal adjustment (about 4 percent of non-oil GDP) would come from a reduction in current expenditure. - Current expenditure would fall by 8 percent in real terms in 2009 (Table 3). - Part of the reduction in current expenditure would be endogenous: reductions in explicit fuel subsidies given lower international fuel prices (examples: Bolivia, Iran, Yemen) and reduced oil revenue coparticipation across various levels of government (examples: Nigeria, Sudan, Venezuela). - The reduction in revenue coparticipation implies “real” cuts in spending by subnational governments. - Capital expenditure in lower income countries would rise on average by 2 percent in real terms. - Table 3. OPCs: Expenditure ratios by income groups (changes 2008-09, in percent of non-oil GDP) - High: Primary spending 1.5% - o/w capital spending 3.3% - o/w current primary spending -1.8% - Upper-middle: Primary spending -1.4% - o/w capital spending -1.1% - o/w current primary spending -0.3% - Lower-middle: Primary spending -4.1% - o/w capital spending -0.6% - o/w current primary spending -3.5% - Low: Primary spending -7.0% - o/w capital spending -0.6% - o/w current primary spending -6.4% ### Prospects for 2010 based on futures-implied oil prices - IMF WEO is projecting a recovery in the average annual oil price for 2010 to US$77 per barrel based on observed future prices. - Preliminary projections for 2010 based on those prices suggest a recovery of the overall fiscal balance by 4 percent of GDP on average. - This improvement would be more marked for high and low income OPCs. - Most of the improvement would come from higher oil revenue, but non-oil revenue ratios would also recover slightly in higher income countries together with global economic activity. - Primary spending ratios to non-oil GDP would, on average, fall slightly relative to 2009 levels. - This would assume some reshuffling within current expenditure items as intergovernmental transfers and fuel subsidies would rise together with oil prices. - Note on Mexico (from text footnote): - Mexico avoided a sharper fiscal adjustment because it could prefinance larger gross borrowing requirements and hedged oil export revenue in 2009 at US$70/barrel through a put option (that protects against downside risks). - Its hedging program has been in place since 1991. This particular operation cost the government US$1.5 billion. - There is no counterparty risk as this operation is collateralized by US Treasuries. ### Fiscal policy in OPCs over the economic cycle: definitions and methodology - Context: - Fiscal policy analysis over the cycle is critical given quasi-fixed exchange rate regimes and high international capital mobility in OPCs, which limit monetary policy and make fiscal policy the main short-term macroeconomic management tool. - Definitions: - Fiscal policy in OPCs has often been defined as expansionary/contractionary when the change in the non-oil primary balance (NOPB) is negative/positive. - Changes in the NOPB can be decomposed: - ∆NOPB = ∆CNOPB + ∆CANOPB - CNOPB: cyclical non-oil primary balance - CANOPB: cyclically adjusted (structural) non-oil primary balance - Fiscal impulse (FI) = change in the cyclically adjusted non-oil primary deficit (∆CANOPB). - Automatic stabilizer (AS) = change in the cyclical non-oil primary deficit (∆CNOPB). - The authors redefine fiscal policy as expansionary/contractionary when the change in the CANOPB is negative/positive, focusing on policy actions under policymakers’ control. - Cyclicality classification: - Expansionary/contractionary fiscal policy when the change in the output gap is positive/negative is called procyclical. - Expansionary/contractionary fiscal policy when the change in the output gap is negative/positive is called countercyclical. - Output gap measurement: - Output gap (OG) = (Y – Yp) / Yp, where Y is actual non-oil real GDP and Yp is potential non-oil real GDP. - Potential non-oil real GDP is estimated using an HP filter. - Fiscal impulses are estimated following Horton, Kumar and Mauro (2009) and Fedelino, Ivanova, and Horton (2009) as changes in the cyclically-adjusted non-oil primary deficit measured in relation to potential (non-oil) output. ### Empirical finding on cyclicality (summary from text) - Fiscal policy was generally procyclical during the boom period (2003-2008). - The authors propose to estimate the change in the output gaps and the fiscal impulses during the boom of 2003-2008 and contrast it with the downturn of 2009; Figure 8 (referenced) shows the median change in the output gap and the fiscal impulses for each subgroup of countries during 2003- (text cuts off). *Source: _wp1028 - 2009. Oil prices hovered around US$40 per barrel for a couple of months early in the year,* ### 2008. We see that the change in the output gap was positive in all subgroups but was more ### _wp1028 - 2008. We see that the change in the output gap was positive in all subgroups but was more ### Fiscal procyclicality: empirical findings - Change in the output gap was positive in all subgroups during 2003-2008 and was more pronounced in upper middle income (UMI) countries. - Fiscal impulses were positive in all subgroups during 2003-2008 and especially large in UMI countries. - Combination of positive changes in output gaps with positive fiscal impulses implies a procyclical fiscal policy response. - Degree of procyclicality (measured by the ratio of the fiscal impulse and the change in the output gap) is higher for low income (LI) countries and lower for high income (HI) countries. - Reverse causality caveat: literature typically assumes output shocks drive fiscal policy; some authors (e.g., Rigobon (2004)) claim fiscal policy shocks drive output. Ilzetzki and Vegh (2008) find causality goes in both directions and evidence of procyclical fiscal policy in developing countries is robust to endogeneity considerations. - HP filter specification: smoothing parameter lambda = 100 for all countries; IMF WEO output projections used until 2014 to reduce end-point sensitivity. - Elasticity assumptions: elasticities of non-oil revenue and non-oil primary expenditure with respect to the output gap are one and zero, respectively. ### Fiscal response during the 2009 downturn - Change in the output gap was negative in all subgroups in 2009 and was more substantial in UMI countries. - Fiscal impulses were negative in all subgroups except in HI countries, where on average fiscal policy was neutral. - Negative changes in output gaps coupled with negative fiscal impulses imply a procyclical fiscal policy response in 2009. - Degree of procyclicality remained higher for LI countries. - Comparative pattern: procyclical response of LI countries during 2009 was stronger than during 2003-2008; for other subgroups the procyclical response was stronger during 2003-2008 than in 2009. - Preliminary projections for 2010 suggest a procyclical fiscal policy stance for all country groups, with some fiscal tightening accompanying negative changes in output gaps; all country groups would display relatively similar degrees of procyclicality in 2010. ### Quantitative measure of procyclicality (Table 4) - Table 4 reports: ΔOG = change in output gap (in percent); FI = fiscal impulse (in percent of non-oil GDP); FI/ΔOG. - Reported figures (2003-2008 and 2009, by income group): - HI: 7.0 7.8 1.1 -3.3 0.0 0.0 - UM: 10.3 13.6 1.3 -7.1 -4.6 0.7 - LM: 3.8 11.7 3.1 -2.1 -4.5 2.1 - L: 2.6 9.0 3.4 -1.6 -8.0 5.1 ### Vulnerability to declines in oil prices: approach and key results - Purpose: sensitivity analysis to assess fiscal vulnerability of OPCs to falls in oil prices. - Simplified linear approach: linear adjustment of oil revenue and oil GDP in the 2009 projections, keeping non-oil revenues and total spending ratios to non-oil GDP unchanged. Caveats noted: assumes local linearity between oil prices and fiscal oil revenue and abstracts from policy responses (exchange rate movements, non-oil revenue increases, spending cuts, changes in intergovernmental transfers, reductions in fuel subsidies). - Fiscal impact estimate: based on the linear oil revenue adjustment approach, the overall fiscal balance in a subsample of OPCs would fall on average by 3½ percent of GDP in response to a US$10 per barrel lower price than in the baseline. - Starting from 2009 projected figures, 18 out of the 21 countries in the analyzed subsample would record overall fiscal deficits after the shock. The average deficit would increase to 7 percent of GDP. - Historical context: this average deficit would be larger than the one recorded in 1998, when oil prices in real terms recorded their lowest level in the period since 1973 (US$17 per barrel at 2009 prices). - Financing implications: on average, the 18 OPCs in the sample recording overall fiscal deficits would have to use about half of their stock of international reserves plus foreign assets held in oil funds projected to be available at the end of 2008. - Reserve coverage outcomes: import coverage of international reserves would fall below 3 months of imports in 5 countries. If there was no fiscal adjustment and/or a depreciation of the currency and oil prices remained low for an additional year, reserve coverage would fall below the 3 month import coverage threshold in half the countries in the subsample. - Countries considered for the US$10 shock exercise: Algeria, Angola, Azerbaijan, Bolivia, Cameroon, Chad, Congo, Ecuador, Equatorial Guinea, Gabon, Indonesia, Iran, Kazakhstan, Mexico, Nigeria, Russia, Sudan, Trinidad and Tobago, Venezuela, Vietnam, and Yemen. Countries with quite substantial foreign assets were excluded from this analysis. ### Fiscal sustainability: methodology and findings - Key sustainability concept: incorporate exhaustibility of oil reserves and government net financial position. - Ten countries had ratio of total proved reserves to (current) production of less than 25 years by end-2008 (based on British Petroleum proved reserves): specific countries listed in text. - Uncertainty drivers in OPCs: oil reserves, oil prices and volumes, costs of production, government take, sale of gas reserves contingent on contracts and infrastructure, and statistical properties of the oil price. - Sustainability benchmark methodology used in this paper: - Step 1: compute government net wealth as present value of projected future oil revenues (evaluated at oil prices prevailing in the respective year of analysis) plus net government financial assets. - Step 2: derive a consumption/spending path (the NOPD) as an annuity over the remaining production period (reserves-to-production ratio in years) plus 15 years. This differs from the usual perpetuity benchmark and is chosen to be more appropriate for countries with short oil production horizons. - Discounting assumptions and other parameters: annual proved oil reserves from BP; constant real oil prices at year-of-analysis level; 4-year average government take from oil production; interest rate of 3 percent in real terms to discount future oil revenue flows and to calculate annuities. - Interpretation: for countries with short production horizons (lower than 25 years), the methodology implies existing government wealth would be consumed over 1 generation (35-40 years), including 15 years to adjust for oil exhaustion, implying a sharp fiscal adjustment immediately after the end of such generation. - Main empirical finding: most OPCs recorded a deterioration in their sustainability position between 2003 and 2009, mainly because of a larger expansion in their NOPD relative to increases in government net wealth. - Summary indicator: “fiscal sustainability ratio” computed as the ratio of implied sustainable or long-term primary expenditure relative to actual primary expenditure. - Countries recording a fiscal sustainability ratio lower than 1 would have to adjust to reach the sustainable benchmark. - Cross-income variation: - Lower income countries, on average, were short of their sustainability benchmarks in both 2003 and 2009; the position deteriorated in low income countries between 2003 and 2009, but improved in lower-middle income countries (the improvement in lower-middle income countries is driven by the Republic of Congo; excluding Congo, lower-middle income countries would have shown an unchanged fiscal sustainability position). - Higher income countries’ fiscal stances were in line with the implied sustainability benchmark in 2009 (just at the horizontal axis), but deteriorated between 2003 and 2009. *Source: IMF working paper content provided in the supplied PDF excerpt.* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp1028.pdf_