## Executive Summary

## Source details

**Canonical URL:** [Executive Summary](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/spn/2010/_spn1005.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/spn/2010/_spn1005.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/spn/2010/_spn1005.pdf.json)

---

### Key findings and context
- Petroleum product subsidies have again started to rise with the rebound in international prices.
- In 2003, global consumer (pre-tax) subsidies for petroleum products totaled nearly $60 billion.
- Pre-tax global subsidies are projected to reach almost $250 billion in 2010.
- Tax-inclusive subsidies, reflecting suboptimal taxation, are estimated to be much larger—$740 billion in 2010, or 1 percent of global GDP.
- G-20 countries account for over 70 percent of tax-inclusive subsidies, with emerging G-20 countries accounting for a sizable share.
- Halving tax-inclusive subsidies could:
  - Reduce projected fiscal deficits by one-sixth in subsidizing countries.
  - Reduce greenhouse emissions by around 15 percent over the long run.
- Subsidy reform strategies should contain measures to mitigate the impact of higher prices on the poorest groups.

### Defining and measuring subsidies
- A price subsidy is defined as the difference between the price facing consumers or producers and a specified “optimal” benchmark price.
- Two measures are used:
  - Pre-tax subsidies: deviations of consumer prices from marginal supply costs (international price adjusted for transport, distribution, and marketing).
  - Tax-inclusive subsidies: estimated based on alternative optimal tax benchmarks (illustrative thresholds used are $0.30 and $0.40 per liter).
- Benchmark adjustments:
  - International U.S. dollar price at nearest hub + $0.10 per liter for shipping + $0.10 per liter for distribution/retailing for net importers;
  - For net exporters the benchmark is the U.S. dollar price (margins cancel out).

### Recent trends and quantitative estimates
- International price volatility periods considered: (1) surge up to mid-2008; (2) collapse in second half of 2008 and rebound in 2009; (3) developments in 2010.
- Pass-through findings (end-2003 to mid-2008):
  - Around two-thirds of countries failed to fully pass through international gasoline price increases; around one-half failed for diesel; more than three-quarters failed for kerosene.
  - Pass-through was especially low for net oil exporters and for countries that already had subsidies in 2003.
- Tax and subsidy incidence (median outcomes):
  - Median tax levels (US$ per liter): End-2003 = 0.12; Mid-2008 = 0.08; End-2008 = 0.48; Mid-2009 = 0.31.
  - Median tax levels by group in End-2003 and Mid-2009:
    - With tax greater than US$ 0.30 in 2003: 0.52 → 0.75
    - With tax less than US$ 0.30 in 2003: 0.11 → 0.24
    - With subsidy in 2003: -0.08 → 0.12
- Global subsidy aggregates (in nominal billions US$; by period)
  - Pre-tax subsidy: End-2003 = 57; Mid-2008 = 519; Mid-2009 = 136; End-2010 (proj.) = 240.
  - Tax-inclusive subsidy (Tax threshold US$ 0.30 per liter): End-2003 = 406; Mid-2008 = 998; Mid-2009 = 524; End-2010 (proj.) = 742.
  - Tax-inclusive subsidy (Tax threshold US$ 0.40 per liter): End-2003 = 579; Mid-2008 = 1,206; Mid-2009 = 672; End-2010 (proj.) = 965.
- Global subsidy aggregates (in percent of total GDP)
  - Pre-tax subsidy: End-2003 = 0.1; Mid-2008 = 0.7; Mid-2009 = 0.2; End-2010 (proj.) = 0.3.
  - Tax-inclusive subsidy (US$ 0.30): End-2003 = 0.9; Mid-2008 = 1.3; Mid-2009 = 0.8; End-2010 (proj.) = 1.0.
  - Tax-inclusive subsidy (US$ 0.40): End-2003 = 1.3; Mid-2008 = 1.6; Mid-2009 = 1.0; End-2010 (proj.) = 1.3.
- Shares of total pre-tax subsidy (percent of total subsidy)
  - Advanced: End-2003 = 0.2; Mid-2008 = 0.0; Mid-2009 = 0.0; End-2010 (proj.) = 0.0.
  - Emerging: End-2003 = 66.0; Mid-2008 = 70.3; Mid-2009 = 60.5; End-2010 (proj.) = 64.6.
  - Developing: End-2003 = 33.8; Mid-2008 = 29.7; Mid-2009 = 39.5; End-2010 (proj.) = 35.4.
  - G-20 share of pre-tax subsidy: End-2003 = 58.4; Mid-2008 = 58.8; Mid-2009 = 52.0; End-2010 (proj.) = 56.3.
- Shares of tax-inclusive subsidy (Tax threshold US$ 0.30 per liter)
  - By income group (of which):
    - Advanced: End-2003 = 36.7; Mid-2008 = 12.7; Mid-2009 = 26.5; End-2010 (proj.) = 22.8.
    - Emerging: End-2003 = 45.6; Mid-2008 = 43.9; Mid-2009 = 52.4; End-2010 (proj.) = 56.7.
    - Developing: End-2003 = 17.7; Mid-2008 = 23.4; Mid-2009 = 21.1; End-2010 (proj.) = 20.5.
  - G-20 share of tax-inclusive subsidy (US$ 0.30): End-2003 = 75.1; Mid-2008 = 66.3; Mid-2009 = 71.1; End-2010 (proj.) = 71.2.
  - Advanced G-20: End-2003 = 36.5; Mid-2008 = 12.6; Mid-2009 = 26.3; End-2010 (proj.) = 22.7.
  - Emerging G-20: End-2003 = 38.6; Mid-2008 = 56.9; Mid-2009 = 44.8; End-2010 (proj.) = 49.7.
- Shares of tax-inclusive subsidy (Tax threshold US$ 0.40 per liter)
  - By income group (of which):
    - Advanced: End-2003 = 41.0; Mid-2008 = 17.4; Mid-2009 = 31.0; End-2010 (proj.) = 27.5.
    - Emerging: End-2003 = 43.1; Mid-2008 = 60.9; Mid-2009 = 50.5; End-2010 (proj.) = 53.9.
    - Developing: End-2003 = 15.9; Mid-2008 = 21.7; Mid-2009 = 18.6; End-2010 (proj.) = 18.6.
  - G-20 share of tax-inclusive subsidy (US$ 0.40): End-2003 = 77.1; Mid-2008 = 68.6; Mid-2009 = 74.1; End-2010 (proj.) = 73.7.
  - Advanced G-20: End-2003 = 40.9; Mid-2008 = 17.2; Mid-2009 = 30.8; End-2010 (proj.) = 27.4.
  - Emerging G-20: End-2003 = 36.2; Mid-2008 = 53.9; Mid-2009 = 43.3; End-2010 (proj.) = 47.2.

### Policy implications and reform priorities
- Full pass-through of international price changes can yield substantial efficiency gains for both net importers and net exporters by avoiding deadweight losses; the magnitude depends on price elasticities and the size of the price increase.
- Subsidy reform should:
  - Reduce the fiscal burden of petroleum product subsidies.
  - Contribute to efforts to address climate change by removing distortions that favor fossil fuel consumption.
  - Include measures to mitigate the impact of higher prices on the poorest groups (targeted mitigating measures).
  - Promote transparency in price-setting and subsidy reporting.
  - Address domestic vested interests that impede reform.
  - Consider cross-border spillover effects and reform price-setting mechanisms.

### Projected global subsidy increases and fiscal context (Section 12)
- IMF (2009b) projects international crude prices will increase by almost one-third between mid-2009 and end-2010.
- Based on pass-through levels observed between end-2003 and mid-2008:
  - Pre-tax subsidies could increase to $240 billion (or nearly 0.3 percent of global GDP).
  - Tax-inclusive subsidies could increase to between $740 billion and $970 billion (or 1.0–1.3 percent of global GDP), depending on the optimal tax benchmark chosen.
- Fiscal context in subsidizing countries:
  - Of the 58 countries with pre-tax subsidies in 2010, 46 have a projected fiscal deficit in 2010.
  - The deficit is expected to exceed 3 percent of GDP in 27 of these countries.
  - Reducing pre-tax subsidies by one-half would decrease the average projected deficit in these subsidizing countries from 2.1 percent of GDP to an average deficit of 0.8 percent of GDP.
  - Reducing tax-inclusive subsidies by one-half would lower the average deficit in subsidizing countries (94 countries if using the $0.30 benchmark) by about one-sixth, from 6.3 percent of GDP to around 5.3 percent of GDP.
- Geographic composition:
  - Emerging and developing countries account for all of projected pre-tax subsidies and 75 percent of tax-inclusive subsidies.
  - G-20 countries account for 57 percent of global pre-tax subsidies but over 70 percent of tax-inclusive subsidies.
  - Emerging G-20 countries account for about half of tax-inclusive subsidies.

### Environmental implications
- Containing subsidies reduces petroleum consumption and associated greenhouse gas emissions:
  - Reducing subsidies by one-half could reduce greenhouse gas emissions by nearly 5 percent by 2050 (based on IEA (2009) estimates and applied ratios described in text).
  - Reducing tax-inclusive subsidies by one-half would result in emissions reductions of 14–17 percent by 2050.
- Potential gains would be larger if subsidies on other fossil fuels such as coal and natural gas are included.

### Distributional incidence and targeting rationale
- Most benefits from universal petroleum subsidies accrue to high-income households:
  - Gasoline subsidies: over 80 percent of total benefits accrue to the richest 40 percent of households.
  - Diesel subsidies: over 65 percent of benefits accrue to the richest 40 percent of households.
  - Liquefied petroleum gas (LPG) subsidies: over 70 percent of benefits accrue to the richest 40 percent of households.
  - Kerosene consumption is more evenly distributed, but substantial kerosene subsidies still go to high-income households (in Africa, more than 45 percent of kerosene subsidies accrue to the top two income quintiles).
- Figure 3: Distribution of Petroleum Product Subsidies by Income Group (selected region quintile shares)
  - Africa: 7.8, 11.5, 15.3, 21.2, 44.2
  - Latin America: 6.3, 12.6, 18.5, 24.9, 37.7
  - Other Regions: 8.9, 12.3, 15.8, 21.0, 42.0
  - Note: Welfare quintiles are based on per capita household consumption for the listed country sample.

### Obstacles to subsidy reform
- Key obstacles:
  - Weak capacity to target mitigating measures to the poor.
  - Lack of transparency in reporting of subsidies.
  - Opposition by vested interests.
  - Cross-border spillover effects (smuggling, diverted benefits, environmental externalities, effects on global demand).
  - Ad hoc price-setting mechanisms.

### Subsidy reform options and policy recommendations
- A. Targeting mitigating measures
  - Compensating the poor requires delivery systems for compensation; gradual reform may be needed when targeted safety nets are not yet feasible.
  - Possible initial steps:
    - Temporarily maintain universal subsidies on commodities more important in poor households’ budgets (e.g., kerosene), acknowledging limits due to market distortions (redirection, smuggling).
    - Introduce short-term mitigation packages: expand existing programs quickly (school meals, education and health user fees, subsidized mass urban transport, cash transfers to vulnerable groups, subsidies for water/electricity below specified thresholds).
    - Identify and protect high-priority public expenditures that benefit the poor (education, health, roads, electrification).
  - Improve safety-net design over time using targeting by socioeconomic and demographic characteristics (categorical), geographic targeting, or self-targeting conditionalities (work or schooling requirements).
  - Pace reforms according to political/administrative capacity and fiscal/macro considerations (risks to growth and inflation).

- B. Promoting transparency
  - Record petroleum subsidies transparently in government accounts; incorporate fiscal-consequential subsidies into the budget on a gross basis and explicitly identify them.
  - Identify and record off-budget subsidies in separate accounts; describe and evaluate arrangements where oil companies provide implicit subsidies without budget support.
  - Measure and analyze subsidies that have no fiscal cost and report them where appropriate (e.g., export bans creating consumer price subsidies).
  - Examples of national reporting systems: Germany’s Biannual Subsidy Reports; Switzerland’s online subsidy database.

- C. Overcoming vested interests
  - Use public information campaigns to inform consumers and taxpayers about subsidy drawbacks (smuggling, shortages, black market activity, corruption) and benefits of reform.
  - Examples:
    - Ghana (2005): used Poverty and Social Impact Analysis to demonstrate regressivity of subsidies and broadcast parliamentary discussions to prepare public.
    - Indonesia (2005 and 2009): linked savings from price increases to cash compensation programs for the poor.
    - Ghana: linked reform to increased resources for high-priority social spending.

- D. Addressing cross-border spillover effects
  - Recognize multiple spillovers: altered trade patterns, smuggling, pass-through to export prices, environmental spillovers, effects on global demand through lack of domestic pass-through.
  - Multilateral cooperation is desirable, particularly where externalities are global (climate change) and unilateral action faces free-rider problems.

- E. Reforming price-setting mechanisms
  - Move away from government direct control of domestic petroleum prices; reform pricing mechanisms to prevent re-emergence of subsidies.
  - First-best: liberalize petroleum prices to depoliticize pricing; liberalization may require:
    - Preparing for concession constraints, liberalizing import/distribution activities, and strengthening regulatory frameworks to detect anti-competitive behavior.
  - If markets are imperfect or price volatility is a concern, implement automatic pricing mechanisms that adjust regularly to international prices, using smoothing features (moving averages, price adjustment caps/triggers, price bands).
  - Countries that have used formula-based mechanisms include Bolivia, Chile, Republic of Congo, Dominica, Ghana, Gabon, Pakistan, Peru, South Africa, and Sri Lanka; however, formulas have sometimes been suspended due to opposition.

### Country experiences with mitigation (Box 3)
- Mozambique (petroleum product prices increased by 38 percent in 2008):
  - Increased budgetary allocations to social protection programs (Direct Social Support, Social Benefits Through Work, Income Generation and Community Development).
  - Increased Food Subsidy Program cash benefits: minimum from 70 MT to 100 MT; maximum from 140 MT to 300 MT.
  - Increased National Institute for Social Protection branches from 19 to 30.
- Indonesia (domestic petroleum product prices increased more than doubled in March and October 2005; prices increased by 25–33 percent in May 2008):
  - Temporary cash transfer program in 2005 aided 19 million poor families using existing databases; similar program in 2008.
  - Some subsidy savings reallocated to education, health, and infrastructure programs.
  - Reduced kerosene subsidy alongside program to increase LPG use.
- Jordan (gradual reduction culminating in full price liberalization in February 2008 with product price increases of 33–76 percent):
  - Increased minimum wage; higher wage increases for low-paid government employees.
  - One-time bonus to low-income government employees and pensioners.
  - Maintained electricity lifeline tariff at current low levels.
  - Provided cash transfers to low-income households whose heads were nongovernmental workers or pensioners.
  - Announced plan to increase funding to the National Aid Fund and improve national safety net with World Bank assistance.

*Source: IMF staff estimates, Executive Summary (SPN/10/05).*

### Executive Summary ......................................................................................................

### Executive Summary

### Key findings and context
- Petroleum product subsidies have again started to rise with the rebound in international prices.
- In 2003, global consumer (pre-tax) subsidies for petroleum products totaled nearly $60 billion.
- Pre-tax global subsidies are projected to reach almost $250 billion in 2010.
- Tax-inclusive subsidies, reflecting suboptimal taxation, are estimated to be much larger—$740 billion in 2010, or 1 percent of global GDP.
- G-20 countries account for over 70 percent of tax-inclusive subsidies, with emerging G-20 countries accounting for a sizable share.
- Halving tax-inclusive subsidies could:
  - Reduce projected fiscal deficits by one-sixth in subsidizing countries.
  - Reduce greenhouse emissions by around 15 percent over the long run.
- Subsidy reform strategies should contain measures to mitigate the impact of higher prices on the poorest groups.

### Defining and measuring subsidies
- A price subsidy is defined as the difference between the price facing consumers or producers and a specified “optimal” benchmark price.
- Two measures are used:
  - Pre-tax subsidies: deviations of consumer prices from marginal supply costs (international price adjusted for transport, distribution, and marketing).
  - Tax-inclusive subsidies: estimated based on alternative optimal tax benchmarks (illustrative thresholds used are $0.30 and $0.40 per liter).
- Benchmark adjustments: international U.S. dollar price at nearest hub + $0.10 per liter for shipping + $0.10 per liter for distribution/retailing for net importers; for net exporters the benchmark is the U.S. dollar price (margins cancel out).

### Recent trends and quantitative estimates
- International price volatility periods considered: (1) surge up to mid-2008; (2) collapse in second half of 2008 and rebound in 2009; (3) developments in 2010.
- Pass-through findings (end-2003 to mid-2008):
  - Around two-thirds of countries failed to fully pass through international gasoline price increases; around one-half failed for diesel; more than three-quarters failed for kerosene.
  - Pass-through was especially low for net oil exporters and for countries that already had subsidies in 2003.
- Tax and subsidy incidence (median outcomes):
  - Median tax levels (US$ per liter): End-2003 = 0.12; Mid-2008 = 0.08; End-2008 = 0.48; Mid-2009 = 0.31.
  - Median tax levels by group in End-2003 and Mid-2009:
    - With tax greater than US$ 0.30 in 2003: 0.52 → 0.75
    - With tax less than US$ 0.30 in 2003: 0.11 → 0.24
    - With subsidy in 2003: -0.08 → 0.12
- Global subsidy aggregates (in nominal billions US$; by period)
  - Pre-tax subsidy: End-2003 = 57; Mid-2008 = 519; Mid-2009 = 136; End-2010 (proj.) = 240.
  - Tax-inclusive subsidy (Tax threshold US$ 0.30 per liter): End-2003 = 406; Mid-2008 = 998; Mid-2009 = 524; End-2010 (proj.) = 742.
  - Tax-inclusive subsidy (Tax threshold US$ 0.40 per liter): End-2003 = 579; Mid-2008 = 1,206; Mid-2009 = 672; End-2010 (proj.) = 965.
- Global subsidy aggregates (in percent of total GDP)
  - Pre-tax subsidy: End-2003 = 0.1; Mid-2008 = 0.7; Mid-2009 = 0.2; End-2010 (proj.) = 0.3.
  - Tax-inclusive subsidy (US$ 0.30): End-2003 = 0.9; Mid-2008 = 1.3; Mid-2009 = 0.8; End-2010 (proj.) = 1.0.
  - Tax-inclusive subsidy (US$ 0.40): End-2003 = 1.3; Mid-2008 = 1.6; Mid-2009 = 1.0; End-2010 (proj.) = 1.3.
- Shares of total pre-tax subsidy (percent of total subsidy)
  - Advanced: End-2003 = 0.2; Mid-2008 = 0.0; Mid-2009 = 0.0; End-2010 (proj.) = 0.0.
  - Emerging: End-2003 = 66.0; Mid-2008 = 70.3; Mid-2009 = 60.5; End-2010 (proj.) = 64.6.
  - Developing: End-2003 = 33.8; Mid-2008 = 29.7; Mid-2009 = 39.5; End-2010 (proj.) = 35.4.
  - G-20 share of pre-tax subsidy: End-2003 = 58.4; Mid-2008 = 58.8; Mid-2009 = 52.0; End-2010 (proj.) = 56.3.
- Shares of tax-inclusive subsidy (Tax threshold US$ 0.30 per liter)
  - By income group (of which):
    - Advanced: End-2003 = 36.7; Mid-2008 = 12.7; Mid-2009 = 26.5; End-2010 (proj.) = 22.8.
    - Emerging: End-2003 = 45.6; Mid-2008 = 43.9; Mid-2009 = 52.4; End-2010 (proj.) = 56.7.
    - Developing: End-2003 = 17.7; Mid-2008 = 23.4; Mid-2009 = 21.1; End-2010 (proj.) = 20.5.
  - G-20 share of tax-inclusive subsidy (US$ 0.30): End-2003 = 75.1; Mid-2008 = 66.3; Mid-2009 = 71.1; End-2010 (proj.) = 71.2.
  - Advanced G-20: End-2003 = 36.5; Mid-2008 = 12.6; Mid-2009 = 26.3; End-2010 (proj.) = 22.7.
  - Emerging G-20: End-2003 = 38.6; Mid-2008 = 56.9; Mid-2009 = 44.8; End-2010 (proj.) = 49.7.
- Shares of tax-inclusive subsidy (Tax threshold US$ 0.40 per liter)
  - By income group (of which):
    - Advanced: End-2003 = 41.0; Mid-2008 = 17.4; Mid-2009 = 31.0; End-2010 (proj.) = 27.5.
    - Emerging: End-2003 = 43.1; Mid-2008 = 60.9; Mid-2009 = 50.5; End-2010 (proj.) = 53.9.
    - Developing: End-2003 = 15.9; Mid-2008 = 21.7; Mid-2009 = 18.6; End-2010 (proj.) = 18.6.
  - G-20 share of tax-inclusive subsidy (US$ 0.40): End-2003 = 77.1; Mid-2008 = 68.6; Mid-2009 = 74.1; End-2010 (proj.) = 73.7.
  - Advanced G-20: End-2003 = 40.9; Mid-2008 = 17.2; Mid-2009 = 30.8; End-2010 (proj.) = 27.4.
  - Emerging G-20: End-2003 = 36.2; Mid-2008 = 53.9; Mid-2009 = 43.3; End-2010 (proj.) = 47.2.

### Policy implications and reform priorities
- Full pass-through of international price changes can yield substantial efficiency gains for both net importers and net exporters by avoiding deadweight losses; the magnitude depends on price elasticities and the size of the price increase.
- Subsidy reform should:
  - Reduce the fiscal burden of petroleum product subsidies.
  - Contribute to efforts to address climate change by removing distortions that favor fossil fuel consumption.
  - Include measures to mitigate the impact of higher prices on the poorest groups (targeted mitigating measures).
  - Promote transparency in price-setting and subsidy reporting.
  - Address domestic vested interests that impede reform.
  - Consider cross-border spillover effects and reform price-setting mechanisms.

*Source: IMF staff estimates, Executive Summary (SPN/10/05).*

### 12. Global subsidies are projected to increase substantially in 2010. Based on

### _spn1005 - 12. Global subsidies are projected to increase substantially in 2010. Based on

### Projected global subsidy increases and fiscal context
- IMF (2009b) projects international crude prices will increase by almost one-third between mid-2009 and end-2010.
- Based on pass-through levels observed between end-2003 and mid-2008:
  - Pre-tax subsidies could increase to $240 billion (or nearly 0.3 percent of global GDP).
  - Tax-inclusive subsidies could increase to between $740 billion and $970 billion (or 1.0–1.3 percent of global GDP), depending on the optimal tax benchmark chosen.
- Fiscal context in subsidizing countries:
  - Of the 58 countries with pre-tax subsidies in 2010, 46 have a projected fiscal deficit in 2010.
  - The deficit is expected to exceed 3 percent of GDP in 27 of these countries.
  - Reducing pre-tax subsidies by one-half would decrease the average projected deficit in these subsidizing countries from 2.1 percent of GDP to an average deficit of 0.8 percent of GDP.
  - Reducing tax-inclusive subsidies by one-half would lower the average deficit in subsidizing countries (94 countries if using the $0.30 benchmark) by about one-sixth, from 6.3 percent of GDP to around 5.3 percent of GDP.
- Geographic composition:
  - Emerging and developing countries account for all of projected pre-tax subsidies and 75 percent of tax-inclusive subsidies.
  - G-20 countries account for 57 percent of global pre-tax subsidies but over 70 percent of tax-inclusive subsidies.
  - Emerging G-20 countries account for about half of tax-inclusive subsidies.

### Environmental implications
- Containing subsidies reduces petroleum consumption and associated greenhouse gas emissions:
  - Reducing subsidies by one-half could reduce greenhouse gas emissions by nearly 5 percent by 2050 (based on IEA (2009) estimates and applied ratios described in text).
  - Reducing tax-inclusive subsidies by one-half would result in emissions reductions of 14–17 percent by 2050.
- Potential gains would be larger if subsidies on other fossil fuels such as coal and natural gas are included.

### Distributional incidence and targeting rationale
- Most benefits from universal petroleum subsidies accrue to high-income households:
  - Gasoline subsidies: over 80 percent of total benefits accrue to the richest 40 percent of households.
  - Diesel subsidies: over 65 percent of benefits accrue to the richest 40 percent of households.
  - Liquefied petroleum gas (LPG) subsidies: over 70 percent of benefits accrue to the richest 40 percent of households.
  - Kerosene consumption is more evenly distributed, but substantial kerosene subsidies still go to high-income households (in Africa, more than 45 percent of kerosene subsidies accrue to the top two income quintiles).
- Figure 3: Distribution of Petroleum Product Subsidies by Income Group (selected region quintile shares)
  - Africa: 7.8, 11.5, 15.3, 21.2, 44.2
  - Latin America: 6.3, 12.6, 18.5, 24.9, 37.7
  - Other Regions: 8.9, 12.3, 15.8, 21.0, 42.0
  - Note: Welfare quintiles are based on per capita household consumption for the listed country sample.

### Obstacles to subsidy reform
- Key obstacles:
  - Weak capacity to target mitigating measures to the poor.
  - Lack of transparency in reporting of subsidies.
  - Opposition by vested interests.
  - Cross-border spillover effects (smuggling, diverted benefits, environmental externalities, effects on global demand).
  - Ad hoc price-setting mechanisms.

### Subsidy reform options and policy recommendations
- A. Targeting mitigating measures
  - Compensating the poor requires delivery systems for compensation; gradual reform may be needed when targeted safety nets are not yet feasible.
  - Possible initial steps:
    - Temporarily maintain universal subsidies on commodities more important in poor households’ budgets (e.g., kerosene), acknowledging limits due to market distortions (redirection, smuggling).
    - Introduce short-term mitigation packages: expand existing programs quickly (school meals, education and health user fees, subsidized mass urban transport, cash transfers to vulnerable groups, subsidies for water/electricity below specified thresholds).
    - Identify and protect high-priority public expenditures that benefit the poor (education, health, roads, electrification).
  - Improve safety-net design over time using targeting by socioeconomic and demographic characteristics (categorical), geographic targeting, or self-targeting conditionalities (work or schooling requirements).
  - Pace reforms according to political/administrative capacity and fiscal/macro considerations (risks to growth and inflation).

- B. Promoting transparency
  - Record petroleum subsidies transparently in government accounts; incorporate fiscal-consequential subsidies into the budget on a gross basis and explicitly identify them.
  - Identify and record off-budget subsidies in separate accounts; describe and evaluate arrangements where oil companies provide implicit subsidies without budget support.
  - Measure and analyze subsidies that have no fiscal cost and report them where appropriate (e.g., export bans creating consumer price subsidies).
  - Examples of national reporting systems: Germany’s Biannual Subsidy Reports; Switzerland’s online subsidy database.

- C. Overcoming vested interests
  - Use public information campaigns to inform consumers and taxpayers about subsidy drawbacks (smuggling, shortages, black market activity, corruption) and benefits of reform.
  - Examples:
    - Ghana (2005): used Poverty and Social Impact Analysis to demonstrate regressivity of subsidies and broadcast parliamentary discussions to prepare public.
    - Indonesia (2005 and 2009): linked savings from price increases to cash compensation programs for the poor.
    - Ghana: linked reform to increased resources for high-priority social spending.

- D. Addressing cross-border spillover effects
  - Recognize multiple spillovers: altered trade patterns, smuggling, pass-through to export prices, environmental spillovers, effects on global demand through lack of domestic pass-through.
  - Multilateral cooperation is desirable, particularly where externalities are global (climate change) and unilateral action faces free-rider problems.

- E. Reforming price-setting mechanisms
  - Move away from government direct control of domestic petroleum prices; reform pricing mechanisms to prevent re-emergence of subsidies.
  - First-best: liberalize petroleum prices to depoliticize pricing; liberalization may require:
    - Preparing for concession constraints, liberalizing import/distribution activities, and strengthening regulatory frameworks to detect anti-competitive behavior.
  - If markets are imperfect or price volatility is a concern, implement automatic pricing mechanisms that adjust regularly to international prices, using smoothing features (moving averages, price adjustment caps/triggers, price bands).
  - Countries that have used formula-based mechanisms include Bolivia, Chile, Republic of Congo, Dominica, Ghana, Gabon, Pakistan, Peru, South Africa, and Sri Lanka; however, formulas have sometimes been suspended due to opposition.

### Country experiences with mitigation (Box 3)
- Mozambique (petroleum product prices increased by 38 percent in 2008):
  - Increased budgetary allocations to social protection programs (Direct Social Support, Social Benefits Through Work, Income Generation and Community Development).
  - Increased Food Subsidy Program cash benefits: minimum from 70 MT to 100 MT; maximum from 140 MT to 300 MT.
  - Increased National Institute for Social Protection branches from 19 to 30.
- Indonesia (domestic petroleum product prices increased more than doubled in March and October 2005; prices increased by 25–33 percent in May 2008):
  - Temporary cash transfer program in 2005 aided 19 million poor families using existing databases; similar program in 2008.
  - Some subsidy savings reallocated to education, health, and infrastructure programs.
  - Reduced kerosene subsidy alongside program to increase LPG use.
- Jordan (gradual reduction culminating in full price liberalization in February 2008 with product price increases of 33–76 percent):
  - Increased minimum wage; higher wage increases for low-paid government employees.
  - One-time bonus to low-income government employees and pensioners.
  - Maintained electricity lifeline tariff at current low levels.
  - Provided cash transfers to low-income households whose heads were nongovernmental workers or pensioners.
  - Announced plan to increase funding to the National Aid Fund and improve national safety net with World Bank assistance.

*International Monetary Fund. Excerpt from IMF Staff Position Note chapter on fuel and food price subsidies and reform options.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/spn/2010/_spn1005.pdf_