## _tnm1607 - EXECUTIVE SUMMARY

## Source details

**Canonical URL:** [_tnm1607 - EXECUTIVE SUMMARY](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/tnm/2016/_tnm1607.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/tnm/2016/_tnm1607.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/tnm/2016/_tnm1607.pdf.json)

---

### Introduction: rationale and context
- Post-tax fuel subsidies were estimated at US$5.3 trillion in 2015 (6.5 percent of global GDP).
- Recent decline in international oil prices has reduced fuel subsidies, but post-tax subsidies remain significant in many countries.
- Fuel subsidies are characterized as inefficient and regressive:
  - distort resource allocation by encouraging excessive fuel consumption and favoring capital-intensive industries;
  - exert pressures on the balance of payments for net oil importers and promote smuggling where domestic prices differ;
  - create a high fiscal burden that can crowd out education, health, and other social expenditures;
  - disproportionately benefit higher-income households, widening income inequalities.
- Concern about impacts on low-income households often blocks subsidy reform, making robust incidence analysis crucial for policy design and protection measures.

### Methodology — theoretical framework and transmission channels
- Direct and indirect channels:
  - Direct channel: households purchase fuel for cooking, heating, lighting, and private transport.
  - Indirect channel: households buy goods and services produced using fuel inputs.
- Direct welfare impact (money measure of decline in real income Y) under unchanged demand:
  - dY/Y = b_i * dP_i/P_i
  - For multiple fuel products: dY/Y = Σ_{i=1}^{m} b_i * dP_i/P_i
  - Definitions: b_i is the share of spending on fuel product i in total household expenditure; dP_i/P_i is the percentage change in the price of fuel product i.
  - Numerical example preserved: if diesel accounts for 10 percent of total household spending, a 10 percent increase in the diesel price implies a 1 percent decline in household welfare.
- Indirect welfare impact:
  - dY/Y = Σ_{j=1}^{n} b_j * dP_j/P_j
  - Definitions: b_j is the budget share of good or service j (non-fuel); dP_j/P_j is the corresponding price change following fuel price changes.
- Price-shifting model (Coady and Newhouse, 2006) to estimate pass-through:
  - Production technology captured by input-output coefficient matrix A, with a_{ij} denoting cost of input i in producing one unit of output j.
  - Simplified relation: ∆P_other = ∆P_fuel * (I − A)^{-1}
- Key assumptions and limitations:
  - Factor prices constant; only exogenous price changes are in the petroleum sector.
  - Model does not account for changes in consumption and production patterns → provides an upper-bound (first-round) welfare impact estimate.
  - Does not capture production-structure changes (e.g., improvements in fuel efficiency) or substitution effects.

### A new Excel-based tool: features and data requirements
- Purpose: distributional incidence analysis of fuel subsidy reform via an Excel template.
- Key features:
  - Requires limited inputs and uses commonly available software (Excel).
  - Combines household survey data with input-output tables to estimate:
    - direct and indirect welfare impacts of a fuel price increase;
    - distribution of impacts across income groups;
    - share of fuel subsidies accruing to different income groups;
    - price increases in other sectors following higher fuel prices.
  - Designed to be straightforward to set up and apply with commonly available data.
- Data requirements:
  - Household survey: household characteristics (size, sampling weight, location, total expenditure) and consumption by categories; for fuel products, consumption for each fuel product is desirable.
  - Input-output matrix: derived shares of different inputs in production cost structure by sector.
- Template workflow (steps preserved):
  1. "Input" tab — specify path to household database and input-output matrix and upload; generates drop-down lists.
  2. On same tab — select household consumption variables for fuel products, indicate price changes to simulate; specify whether survey data are at household level or aggregated by income groups; select household characteristic variables and income group definitions; select controlled sectors from input-output sectors.
     - When survey data are at household level, expenditure data should be in nominal terms; when aggregated by groups, expenditure data should be budget shares.
     - If sampling weights or geographic location missing, template has fallback behaviors (equal household weighting; results without urban/rural breakdown).
  3. "Mapping" tab — assign household goods/services to input-output sectors to calculate indirect effects.
  4. "Run" tab — run macros to compute direct, indirect, and total welfare impacts; separate macro available for direct welfare impact (useful if no input-output matrix).
- Handling fuel-sector aggregation cases:
  - Case 1: Each fuel product separate in input-output → ideal.
  - Case 2: Composite fuel sector → user inputs share of each fuel product in intermediate consumption to compute average fuel price increase (default assumes diesel only, changeable with country data).
  - Case 3: Fuel sector mixed with utilities (electricity, water) → user indicates share of fuel products in the aggregate sector and the share of each fuel product in intermediate consumption.
  - When a fuel sector share parameter is required, no default value is provided.
- Outputs:
  - Direct welfare impact by income group with breakdown by fuel product.
  - Share of direct fuel subsidies accruing to each income group (also by fuel product).
  - Price increases in other sectors reflecting their energy intensity; distribution of indirect welfare effect by income group and sectoral contributions.
  - Share of indirect fuel subsidies accruing to each income group.
  - Distribution of total welfare impact and subsidy share by income groups; relative contribution of direct vs. indirect effects.
  - First-round effect on consumer price index (weighted average of price increases for fuel and non-fuel products using household consumption weights).
- In absence of input-output table:
  - Template can estimate only the direct welfare impact by running steps 1 and 2 and executing only the direct-impact macro.
  - Alternative: use indirect-effect information from similar country sources (e.g., Coady, Flamini, and Sears (2015)).

### Illustrative application: Brazil case study (summary)
- Data and setup:
  - Household consumption data: 2005 household survey covering about 56,000 households.
  - Input-output: condensed 2005 input-output table covering 13 sectors (with use of a more disaggregated 110-sector table to isolate fuel sector data combined with 12 aggregated sectors).
  - Simulation: 10 percent increase in gasoline and diesel prices.
  - Controlled sectors for indirect effects in simulation: (i) electricity, gas and water, and (ii) public administration, health and education services.
- Background facts (selected):
  - Petrobras historically had upstream and downstream monopoly; retail prices were government-controlled.
  - An oil stabilization fund existed; rising import costs depleted the fund; accumulated contingent liabilities to Petrobras were only partially covered by a budget transfer of 0.8 percent of 1995 GDP in the mid-1990s.
  - Petroleum law of 1997 liberalized prices; gasoline, diesel and LPG prices began quarterly adjustments from 2001; refinery/downstream sector opened to private sector.
  - Petrobras remains dominant in refining and imports; downstream losses estimated at a cumulative 2.5 percent of GDP during 2011–14.
- Key quantitative simulation results:
  - Direct subsidy distribution:
    - More than half of the direct subsidy on gasoline and diesel is captured by the 20 percent richest households.
    - The poorest quintile receives less than 2 percent of total direct fuel subsidies.
    - Consumption pattern: richest quintile allocates 3.6 percent of their spending to fuel products, "4½ times" more than the poorest quintile.
    - Direct welfare impact: a 10 percent increase in diesel and gasoline prices → 0.36 percent decline in household welfare for the richest quintile; 0.08 percent decline for the poorest quintile.
    - Average direct welfare loss ≈ 0.24 percent.
    - Note: diesel is not widely consumed by households in Brazil, making the direct effect correspondingly small.
  - Indirect welfare impact:
    - For a 10 percent increase in gasoline and diesel prices, the indirect welfare impact ≈ 0.6 percent decline in household welfare, broadly similar across quintiles.
    - Indirect subsidy share: richest quintile receives about 50 percent of indirect subsidy; lowest quintile about 6 percent.
    - Sectoral composition: poorest quintile more affected by price increases in agriculture (food); richest quintile more affected by manufactured products; food, transport, and manufactured sectors are among the most energy-intensive.
  - Total impact (direct + indirect):
    - Total welfare impact (10 percent fuel price increase) estimated at an average 0.34 percent decline in household welfare.
    - Share of total impact attributable to direct impact: 70 percent.
    - Distributional conclusion: removal of fuel subsidies is progressive — welfare impact increases significantly with income level.
    - Rural households would be slightly more affected than urban households.
    - First-round effect on price level: a 10 percent increase in fuel prices leads to an increase in price level by slightly less than 1 percent.

### Key findings and policy-relevant points
- Core finding: fuel subsidy removal is typically progressive — higher-income households capture a larger share of subsidy benefits, so removing subsidies tends to favor lower-income households in distributional terms.
- The Excel tool provides a rapid, transparent way to quantify:
  - who benefits from existing fuel subsidies (by income group);
  - the magnitude and distribution of welfare losses from price increases (direct and indirect);
  - first-round effects on consumer prices across sectors.
- Incidence analysis should inform policy decisions on how to protect vulnerable households during subsidy removal.

### Conclusion and future work
- The note documents a tractable methodology and a practical Excel tool to support policymakers in evaluating distributional effects of fuel subsidy reform.
- Advantages: limited inputs, Excel-based, easy to use; basic data requirements are a household survey and an input-output table.
- Limitations reiterated: estimates are upper bounds as the tool does not account for changes in consumption and production patterns (substitution and broader general-equilibrium responses).
- Brazil application summary:
  - Richest quintile receives about half of total subsidies; poorest quintile receives less than 2 percent.
  - A 10 percent increase in gasoline and diesel prices → modest household welfare impact (0.34 percent decline on average) and price level impact (about 1 percent).
  - Overall, removal of fuel subsidies is progressive.
- Suggested next steps for tool extension:
  1. Expand template to analyze distributional impact of reforms of food and fertilizer prices.
  2. Extend to reform of public utility tariff structures for electricity and water.
  3. Extend to tax policy reforms, in particular consumption taxes.

*Source: EXECUTIVE SUMMARY, Technical Notes and Manuals 16/07 | 2016*

### EXECUTIVE SUMMARY

### _tnm1607 - EXECUTIVE SUMMARY

### Introduction: rationale and context
- Recent decline in international oil prices has reduced fuel subsidies, but post-tax subsidies remain significant in many countries.
- Post-tax subsidies were estimated at US$5.3 trillion in 2015 (6.5 percent of global GDP).
- Fuel subsidies are characterized as inefficient and regressive; they:
  - distort resource allocation by encouraging excessive fuel consumption and favoring capital-intensive industries;
  - exert pressures on the balance of payments for net oil importers and promote smuggling where domestic prices differ;
  - create a high fiscal burden that can crowd out education, health, and other social expenditures;
  - disproportionately benefit higher-income households, widening income inequalities.
- Concern about impacts on low-income households often blocks subsidy reform, making robust incidence analysis crucial for policy design and protection measures.

### Methodology: theoretical framework and transmission channels
- Impacts of fuel price changes on household welfare occur through:
  - Direct channel: when households purchase fuel for cooking, heating, lighting, and private transport.
  - Indirect channel: when households buy goods and services produced using fuel inputs.
- The overall welfare impact equals the sum of direct and indirect impacts calculated at the household level and aggregated by income groups (quintiles or deciles).
- Direct welfare impact:
  - Money measure of direct loss in welfare (decrease in real income Y) is a function of the budget share of the fuel product.
  - Numerical example preserved from source: if diesel accounts for 10 percent of total household spending, a 10 percent increase in the diesel price implies a 1 percent decline in household welfare.
  - The formulation extends to multiple fuel products by summing across fuel products.
- Indirect welfare impact:
  - Calculated analogously using budget shares of non-fuel goods and services and the price changes in those sectors induced by fuel price changes.
- Note: substitution effects (changes in consumption and production patterns) can reduce the overall welfare impact; these are acknowledged though the baseline formulae assume unchanged demand.

### A new Excel-based tool: features and data requirements
- The note presents a user-friendly, Excel-based template for distributional incidence analysis of fuel subsidy reform.
- Key features:
  - Requires limited inputs and uses commonly available software (Excel).
  - Combines household survey data with input-output tables to estimate:
    - direct and indirect welfare impacts of a fuel price increase;
    - distribution of impacts across income groups;
    - share of fuel subsidies accruing to different income groups;
    - price increases in other sectors following higher fuel prices.
- The tool is designed to be straightforward to set up and apply with commonly available data.

### Illustrative application: Brazil case study (summary)
- Illustration uses Brazil data: the 2005 household survey and input-output table.
- The note includes figures relevant to the Brazil case:
  - International and domestic prices for gasoline and diesel, Jan 2004–Dec 2014.
  - Fuel budget shares, direct subsidy shares and direct welfare impact of fuel price increases.
  - Indirect subsidy shares, price changes in other sectors, and indirect welfare impact.
  - Distribution of total welfare impact, subsidy shares and first-round effect on price level.
- Core empirical finding reinforced by the Brazil example:
  - Fuel subsidies benefit well-off households and their removal would be progressive.

### Key findings and policy-relevant points
- Fuel subsidy removal is typically progressive: higher-income households capture a larger share of subsidy benefits, so removing subsidies tends to favor lower-income households in distributional terms.
- Incidence analysis should inform policy decisions on how to protect vulnerable households during subsidy removal.
- The Excel tool provides a rapid, transparent way to quantify:
  - who benefits from existing fuel subsidies (by income group);
  - the magnitude and distribution of welfare losses from price increases (direct and indirect);
  - first-round effects on consumer prices across sectors.

### Conclusion and future work
- The note documents a tractable methodology and a practical Excel tool to support policymakers in evaluating distributional effects of fuel subsidy reform.
- The tool’s simplicity and limited data needs make it suitable for broad application; the Brazil case serves as an illustrative example.
- Future work implications (implied by structure): refine treatment of substitution effects and broader general-equilibrium responses beyond first-round price impacts.

*Source: EXECUTIVE SUMMARY, Technical Notes and Manuals 16/07 | 2016*

### Section	
  IV	
  concludes.

### Section IV concludes.

### Methodology — Theoretical framework
- Overall welfare impact = sum of direct and indirect impacts calculated at the household level and averaged by income groups (quintiles or deciles).
- Direct welfare impact (money measure / decline in real income Y) under unchanged demand:
  - Formula presented: dY/Y = b_i * dP_i/P_i
  - For multiple fuel products: dY/Y = Σ_{i=1}^{m} b_i * dP_i/P_i
  - Definitions preserved: b_i is the share of spending on fuel product i in total household expenditure; dP_i/P_i is the percentage change in the price of fuel product i.
  - Numerical example: if diesel accounts for 10 percent of total spending, a 10 percent increase in diesel price → a 1 percent decline in household welfare.
- Indirect welfare impact:
  - Formula analogous to the direct impact: dY/Y = Σ_{j=1}^{n} b_j * dP_j/P_j
  - Definitions preserved: b_j is the budget share of good or service j (non-fuel); dP_j/P_j is the corresponding price change following fuel price changes; n is the total number of non-fuel goods and services.
- Price-shifting model (Coady and Newhouse, 2006) to estimate pass-through from fuel price increases to prices of goods and services:
  - Production technology captured by input-output coefficient matrix A, with a_{ij} denoting cost of input i in producing one unit of output j — implies a Leontief production function (firm demand for inputs assumed unaffected by input price changes).
  - Simplified two-category sector formula: ∆P_other = ∆P_fuel * (I − A)^{-1}
  - Assumptions and limitations:
    - Factor prices constant; only exogenous price changes are in the petroleum sector.
    - Model does not account for changes in consumption behavior → provides an upper-bound of welfare impact.
    - Does not capture production-structure changes (e.g., improvements in fuel efficiency).

### A new tool for distributional incidence analysis (Excel template)
- Tool characteristics:
  - Built in Excel, presented as a template with input sheets and graphical/tabled outputs.
  - User friendly, requires limited inputs, does not need expensive software.
  - Data requirements: a household survey database and an input-output matrix.
    - Household survey: household characteristics (size, sampling weight, location, total expenditure) and consumption by categories; for fuel products, consumption for each fuel product is desirable.
    - Input-output matrix: derived shares of different inputs in production cost structure by sector.
- Steps to set up and run the template:
  1. "Input" tab — specify path to household database and input-output matrix and upload; generates drop-down lists.
  2. On same tab — select household consumption variables for fuel products, indicate price changes to simulate; specify whether survey data are at household level or aggregated by income groups; select household characteristic variables and income group definitions; select controlled sectors from input-output sectors.
     - Note: when survey data are at household level, expenditure data should be in nominal terms; when aggregated by groups, expenditure data should be budget shares.
     - If sampling weights or geographic location missing, template has fallback behaviors (equal household weighting; results without urban/rural breakdown).
  3. "Mapping" tab — assign household goods/services to input-output sectors to calculate indirect effects.
  4. "Run" tab — run macros to compute direct, indirect, and total welfare impacts; separate macro available for direct welfare impact (useful if no input-output matrix).
- Handling fuel-sector aggregation cases:
  - Case 1: Each fuel product is separate in input-output → ideal; no extra input required.
  - Case 2: Composite fuel sector → user inputs share of each fuel product in intermediate consumption to compute average fuel price increase (default assumes diesel only, changeable with country data).
  - Case 3: Fuel sector mixed with utilities (electricity, water) → user indicates share of fuel products in the aggregate sector and the share of each fuel product in intermediate consumption.
  - Note: when fuel sector share parameter is required, no default value is provided.
- Outputs of the template include:
  - Direct welfare impact by income group with breakdown by fuel product.
  - Share of direct fuel subsidies accruing to each income group (also by fuel product).
  - Price increases in other sectors reflecting their energy intensity, distribution of indirect welfare effect by income group and sectoral contributions (e.g., food and transport most sensitive).
  - Share of indirect fuel subsidies accruing to each income group.
  - Distribution of total welfare impact and subsidy share by income groups; relative contribution of direct vs. indirect effects to total expenditure effect.
  - First-round effect on consumer price index (weighted average of price increases for fuel and non-fuel products using household consumption weights).
- In absence of input-output table:
  - Template can estimate only the direct welfare impact by running steps 1 and 2 and then executing only the direct-impact macro in step 4.
  - Alternative: use indirect-effect information from similar country sources (e.g., Coady, Flamini, and Sears (2015)).

### Illustrative country-case study: Brazil
- Motivation for Brazil case study:
  - Brazil has one of the highest income inequality levels despite declines.
  - Domestic fuel consumption heavily dependent on fuel imports and has outpaced fuel exports.
  - Availability of rich household database and input-output table.
- Background (selected factual points preserved):
  - Petrobras historically had upstream and downstream monopoly; retail prices were government-controlled.
  - An oil stabilization fund existed; rising import costs depleted the fund; accumulated contingent liabilities to Petrobras were only partially covered by a budget transfer of 0.8 percent of 1995 GDP in the mid-1990s; Petrobras absorbed the remaining loss.
  - Petroleum law of 1997 liberalized prices; gasoline, diesel and LPG prices began quarterly adjustments from 2001; refinery/downstream sector opened to private sector.
  - Petrobras remains dominant in refining and imports; downstream losses estimated at a cumulative 2.5 percent of GDP during 2011–14.
- Simulation setup:
  - Simulated a 10 percent increase in gasoline and diesel prices.
  - Household consumption data from the 2005 household survey covering about 56,000 households.
  - Used a condensed 2005 input-output table covering 13 sectors (with mention that the more disaggregated 110-sector table was used to isolate fuel sector data combined with 12 aggregated sectors).
- Key quantitative simulation results (all numeric values preserved exactly):
  - Direct subsidy distribution:
    - More than half of the direct subsidy on gasoline and diesel is captured by the 20 percent richest households.
    - The poorest quintile receives less than 2 percent of total direct fuel subsidies.
    - Consumption pattern: richest quintile allocates 3.6 percent of their spending to fuel products, "4½ times" more than the poorest quintile.
    - Direct welfare impact: a 10 percent increase in diesel and gasoline prices → 0.36 percent decline in household welfare for the richest quintile; 0.08 percent decline for the poorest quintile.
    - Average direct welfare loss ≈ 0.24 percent.
    - Note: since diesel is not widely consumed by households in Brazil, the direct effect is correspondingly small.
  - Indirect welfare impact:
    - For a 10 percent increase in gasoline and diesel prices, the indirect welfare impact ≈ 0.6 percent decline in household welfare, broadly similar across quintiles.
    - Indirect subsidy share: richest quintile receives about 50 percent of indirect subsidy; lowest quintile about 6 percent.
    - Sectoral composition: poorest quintile more affected by price increases in agriculture (food); richest quintile more affected by manufactured products; food, transport, and manufactured sectors are among the most energy-intensive.
    - Assumption for indirect effects in simulation: two controlled sectors — (i) electricity, gas and water, and (ii) public administration, health and education services.
  - Total impact (direct + indirect):
    - Total welfare impact (10 percent fuel price increase) estimated at an average 0.34 percent decline in household welfare.
    - Share of total impact attributable to direct impact: 70 percent.
    - Distributional conclusion: removal of fuel subsidies is progressive — welfare impact increases significantly with income level.
    - Rural households would be slightly more affected than urban households.
    - First-round effect on price level: a 10 percent increase in fuel prices leads to an increase in price level by slightly less than 1 percent.

### Conclusion and future work
- The note presents an easy and user-friendly Excel tool for incidence analysis of fuel price subsidy reform based on Coady and Newhouse (2006).
- Advantages: limited inputs, Excel-based, easy to use; basic data requirements are a household survey and an input-output table.
- The tool enables analysis of:
  - Distribution of direct, indirect, and total welfare impacts across income groups (quintiles/deciles).
  - Share of fuel subsidies accruing to each income group (direct, indirect, total).
  - Price implications for other economic sectors and their energy intensity.
  - First-round effect on consumer price index.
- Limitations reiterated: estimates are upper bounds as tool does not account for changes in consumption and production patterns.
- Brazil application summary:
  - Richest quintile receives about half of total subsidies; poorest quintile receives less than 2 percent.
  - A 10 percent increase in gasoline and diesel prices → modest household welfare impact (0.34 percent decline on average) and price level impact (about 1 percent).
  - Overall, removal of fuel subsidies is progressive.
- Suggested next steps for tool extension:
  1. Expand template to analyze distributional impact of reforms of food and fertilizer prices.
  2. Extend to reform of public utility tariff structures for electricity and water.
  3. Extend to tax policy reforms, in particular consumption taxes.

*Technical Notes and Manuals 16/07 | 2016*

### References

### References

### Citations
- Araar, Abdelkrim, and Paolo Verme, 2012, “Reforming Subsidies: A Tool-Kit for Policy Simulations,” Policy Research Working Paper Series 6148, The World Bank.
- Berg, Andrew, and Jonathan Ostry, 2011, “Inequality and Unsustainable Growth: Two Sides of the Same Coin?,” IMF Staff Discussion Notes 11/08, Washington: International Monetary Fund.
- Clements, Benedict, David Coady, Stefania Fabrizio, Sanjeev Gupta, Trevor Serge Coleridge Alleyne, and Carlo A. Sdralevich, eds., 2013. Energy Subsidy Reform: Lessons and Implications. Washington: International Monetary Fund
- Coady, David, 2006, “Indirect Tax and Public Pricing Reforms,” in Analyzing the Distributional Impact of Reforms: A Practitioner’s Guide to Pension, Health, Labor Market, Public Sector Downsizing, Taxation, Decentralization and Macroeconomic Modeling, A. Coudouel and S. Paternostro (eds.). Washington: World Bank.
- Coady, David, and David Newhouse, 2006 “Ghana: Evaluating the Fiscal and Social Costs of Increases in Domestic Fuel Prices,” in Poverty and Social Impact Analysis of Reform, Coudouel, A., A. Dani and S. Paternostro (eds.). Washington: World Bank.
- Coady, David, Ian W.H. Parry, Louis Sears, and Baoping Shang, 2015, “How Large are Global Energy Subsidies?,” IMF Working Papers No. 15/105, Washington: International Monetary Fund.
- Coady, David, Valentina Flamini, and Louis Sears, 2015, “The Unequal Benefits of Fuel Subsidies Revisited: Evidence for Developing Countries,” IMF Working Paper No. 15/250, Washington: International Monetary Fund.
- Oliveira, Patrícia, and Edmar Almeida, 2015, “Determinants of Fuel Price Control in Brazil and Price Policy Options”, 5th Latin American Energy Economics Meeting.

### Publication and contact
- TNM/16/07
- International Monetary Fund
- Fiscal Affairs Department
- 700 19th Street NW
- Washington, DC 20431
- USA
- Tel: 1-202-623-8554
- Fax: 1-202-623-6073

*TNM/16/07 International Monetary Fund Fiscal Affairs Department 700 19th Street NW Washington, DC 20431 USA Tel: 1-202-623-8554 Fax: 1-202-623-6073*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/tnm/2016/_tnm1607.pdf_