## wpiea2021271-print-pdf

## Source details

**Canonical URL:** [wpiea2021271-print-pdf](https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021271-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2021/english/wpiea2021271-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2021/english/wpiea2021271-print-pdf.pdf.json)

---

### Summary of major themes and objectives
- Objective: Assess the response of consumer prices to changes in gasoline prices, disaggregating effects across the overall CPI and 12 COICOP categories, and examining distributional impacts across household quintiles and across advanced versus developing economies.
- Data enhancement: Builds on the Global Monthly Retail Fuel Price Database by Kpodar and Abdallah (2017), extending the end period to December 2020, expanding country coverage to 190 countries (from 162), and adding product differentiation (premium and regular gasoline and diesel).
- Motivation: Domestic retail fuel prices—not just crude oil prices—drive household and firm decisions due to taxes, regulation, refinery margins, weather events, and market structure. Using retail prices addresses biases from relying solely on crude oil prices.

### Key empirical findings
- General pass-through
  - Fuel price increases are associated with higher inflation.
  - The magnitude of pass-through is smaller but more persistent in developing countries than in advanced economies.
  - Persistence in developing economies results in a stronger impact of fuel price changes on consumer price levels.
- CPI component responses
  - Transport CPI is the most sensitive to fuel prices in both advanced and developing economies.
  - Developing countries: 10 out of the 12 CPI components show a positive and significant response to gasoline price increases.
  - Advanced economies: 7 out of the 12 CPI components show a positive and significant response.
- Crude oil vs. retail fuel prices (EU example)
  - Fuel taxes account on average for about 60 percent of retail gasoline prices in June 2021 in the EU; less than 20 percent in the US.
  - A 1 percentage point increase in crude oil prices is associated with a 0.015 percentage point increase in consumer price inflation at peak level (EU sample).
  - A 1 percentage point increase in after-tax retail gasoline prices is associated with a 0.055 percentage point increase at peak.
  - Using before-tax retail gasoline price yields a 0.02 percentage point pass-through at peak (close to crude oil result).
  - Conclusion: using crude oil prices significantly understates pass-through because fuel tax dynamics are not captured.
- Distributional effects across households
  - Gasoline price increases reduce purchasing power for all income groups.
  - Short term impact is larger for the richest households than for the poorest households.
  - Advanced economies: differential impact (rich vs. poor) is short-lived (up to 6 months).
  - Developing countries: differential impact persists beyond a year.
- Historical context and trends
  - Crude oil prices dropped to less than $10 dollars a barrel in March 2020 and rose above $80 dollars a barrel in October 2021.
  - A further increase in crude oil prices is considered likely due to global reopening and active supply management by oil producing countries.
- Climate policy context
  - IMF (2021) estimate: a global carbon tax of around $75 per ton is needed to align with the Paris Agreement goal of keeping warming below 2 oC above preindustrial levels.
  - Current global average emissions price is estimated at only $3 per ton.

### Methodology and data approach
- Price series and coverage
  - Uses updated Global Monthly Retail Fuel Price Database, extending previous series through December 2020 and increasing coverage to 190 countries.
  - Adds product differentiation: premium and regular gasoline and diesel prices.
  - Period of data availability now ranges from January 2000 to December 2020 for most countries (a few of them have historical data back to the 70s).
  - The dataset consists of more than 425,000 country-month-product observations.
  - Data are provided in local currencies and also converted in US dollars using the prevailing official exchange rates.
- CPI decomposition and sample
  - Analyzes the overall consumer price index (CPI) and 12 COICOP categories: food, transport, housing and energy, beverages, clothing, communication, education, health, equipment, recreation, hotels and other goods and services.
  - Regression sample: 122 countries (48 advanced/high-income economies and 74 developing economies) with data during January 2000 to June 2019.
- Distributional reconstruction
  - Uses harmonized household survey databases from the World Bank to extract expenditure shares for the poorest quintile (the 20 percent poorest households) and the richest quintile (the 20 percent richest households).
  - Expenditure shares applied over a 15-year window (4 years before the survey year and 10 years afterwards) unless another survey is available.
  - Monthly CPI index calculated for each quintile in 84 countries during January 2000—June 2019.
- Estimation approach
  - Local projection method (Jordà (2005)) augmented following Teulings and Zubanov (2014).
  - Gasoline used as representative fuel product.
  - Model: ∆Ln(CPI)_{i,t+h} regressed on ∆Ln(RFP) and controls; coefficient ϕ_h represents pass-through at horizon h; horizons estimated up to H=11 (12 months).
  - Same framework used to estimate distributional impact by replacing dependent variable with ratio of CPI of richest to poorest quintile.

### Quantitative pass-through results (exact reported estimates)
- Advanced economies
  - A one percentage point increase in retail gasoline price leads to a 0.04 percentage point increase in consumer price inflation at peak level.
  - Pass-through dies out within 2 months after the gasoline price shock.
  - Peak pass-through in advanced economies is double that of developing economies.
- Developing economies
  - A one percentage point increase in gasoline prices pushes consumer prices up by 0.02 percentage point at peak.
  - Pass-through is more persistent: five to six months in developing economies.
  - Cumulative impact on the CPI level is larger in developing countries because persistence more than offsets the lower peak pass-through.
- Contribution to overall inflation
  - Transport, housing and utilities, and food prices account for a combined 60 percent of household expenditure in advanced economies, and 70 percent in developing economies.

### Heterogeneity, channels, and drivers
- Direct channels
  - Mechanical inclusion of fuel prices in CPI components consumed by households (transport, lighting, cooking).
- Indirect channels
  - Cost pass-through via supply chains where fuel is an input to production and transport.
  - Producers may absorb costs (reduce margins, improve efficiency) or pass costs to consumers.
- Market and policy moderating factors
  - Market structure (competition vs. monopoly) influences pass-through extent.
  - Government price regulation can weaken transmission of fuel price increases.
  - Labor market responses and monetary policy reactions (wage pressure, credibility of monetary policy) influence inflation persistence and amplitude.
- Cross-country heterogeneity drivers
  - Energy intensity, labor market flexibility, central bank credibility, share of food and fuel in CPI, past inflation history, and governance quality.
  - Positive oil price shocks tend to have larger effects on inflation than negative shocks (asymmetric response).
- Regional observed patterns
  - Highest gasoline prices observed in Europe; South Asia, Middle East and North Africa, and North America exhibit lower prices.
  - Regional differences in response of retail gasoline prices to the COVID-19 crude oil price drop; consumers in Europe, North and South America, and South Asia benefited from the sharpest drop in Q1 2020.

### Distributional impacts (detailed)
- Metric: Ratio of the CPI index of the richest quintile to the CPI index of the poorest quintile.
  - Positive and significant response of this ratio denotes a progressive impact (CPI for poorest increases less than for richest).
  - Negative and significant response denotes a regressive impact.
- Advanced economies
  - A 1 percent increase in gasoline prices raises the CPI of the richest households by about 0.02 percentage point more than the CPI of the poorest households in the first two months.
  - Marginal impact declines over time to zero from the 6th month onward.
- Developing economies
  - Richer households experience a larger impact, with the differential declining only slightly and persisting in positive territory for more than 12 months.
  - Peak pass-through to inflation increases with household income (richest quintiles experience higher peak pass-through).
  - Regional robustness: progressivity stronger in Asia and Latin America than in Sub-Saharan Africa.

### Robustness and additional controls
- Results remain quantitatively comparable after controlling for:
  - Policy rate of the central bank.
  - Exchange rate movements and international food prices.
- Annex analyses include impulse response functions with additional controls and standardized-variable comparisons for EU countries.

### Policy implications and recommendations
- Inflation management
  - Acknowledge stronger and more persistent pass-through in developing countries when designing monetary and fiscal responses to fuel price shocks.
- Carbon pricing and equity
  - Implementing carbon taxes (e.g., the $75 per ton benchmark) requires measures to mitigate adverse distributional impacts and inflationary consequences, particularly given current global average emissions price of $3 per ton.
- Reform acceptability and social protection
  - Mitigating measures such as targeted transfers or social safety nets are key to offsetting adverse impacts of fuel price increases or subsidy removals on vulnerable households.
  - Evidence suggests partial subsidy removal plus targeted transfers can offset initial income decreases for the poor.
- Structural and institutional reforms to limit persistence
  - Improve energy efficiency.
  - Boost availability of affordable and cleaner energy sources.
  - Improve labor market flexibility.
  - Strengthen monetary policy credibility.
  - Limit price controls.
- Data policy
  - High-frequency, disaggregated retail fuel price data are essential for accurate estimation and policy design; reliance on crude oil prices alone can lead to substantial underestimation of pass-through.

### Annexes, figures, and dataset notes (selected)
- Dataset coverage: sample expanded to 190 countries; dataset now provides premium and regular grades for gasoline and diesel.
- Dataset size: more than 425,000 country-month-product observations.
- Household survey data available for 110 countries; quintile CPIs computed for 84 countries after a minimum 80 percent coverage threshold.
- Annex figures include impulse response functions, regional IRFs, and standardized-variable comparisons; Annex tables provide fixed-effect estimations with panel statistics as reported.

*Source: wpiea2021271-print-pdf - REFERENCES ____________________________________________________________22*

### REFERENCES ____________________________________________________________22

### wpiea2021271-print-pdf - REFERENCES ____________________________________________________________22

### Summary of major themes and objectives
- Objective: Assess the response of consumer prices to changes in gasoline prices, disaggregating effects across the overall CPI and 12 COICOP categories, and examining distributional impacts across household quintiles and across advanced versus developing economies.
- Data enhancement: Builds on the Global Monthly Retail Fuel Price Database by Kpodar and Abdallah (2017), extending the end period to December 2020, expanding country coverage to 190 countries (from 162), and adding product differentiation (premium and regular gasoline and diesel).
- Motivation: Domestic retail fuel prices—not just crude oil prices—drive household and firm decisions due to taxes, regulation, refinery margins, weather events, and market structure. Using retail prices addresses biases from relying solely on crude oil prices.

### Key empirical findings
- General pass-through:
  - Fuel price increases are associated with higher inflation.
  - The magnitude of pass-through is smaller but more persistent in developing countries than in advanced economies.
  - Persistence in developing economies results in a stronger impact of fuel price changes on consumer price levels.
- CPI component responses:
  - Transport CPI is the most sensitive to fuel prices in both advanced and developing economies.
  - Developing countries show broader pass-through: 10 out of the 12 CPI components show a positive and significant response to gasoline price increases.
  - Advanced economies show significant responses in 7 out of the 12 CPI components.
- Crude oil vs. retail fuel prices:
  - For European Union countries, using crude oil prices underestimates pass-through by a large margin because pass-through of fuel taxes is not accounted for when using crude oil prices.
- Distributional effects across households:
  - Gasoline price increases reduce purchasing power for all income groups.
  - The impact is larger for the richest households than for the poorest households in the short term.
  - In advanced economies the differential impact (rich vs. poor) is short-lived (up to 6 months).
  - In developing countries the differential impact persists beyond a year.
- Historical context and trends:
  - Crude oil prices dropped to less than $10 dollars a barrel in March 2020 and rose above $80 dollars a barrel in October 2021.
  - A further increase in crude oil prices is considered likely due to global reopening and active supply management by oil producing countries.
- Climate policy context:
  - IMF (2021) estimate: a global carbon tax of around $75 per ton is needed to align with the Paris Agreement goal of keeping warming below 2 oC above preindustrial levels.
  - Current global average emissions price is estimated at only $3 per ton, far below the $75 per ton benchmark.
  - Distributional and inflation consequences of higher fuel prices are central to policy design for carbon taxation and related reforms.

### Methodology and data approach (as described)
- Price series and coverage:
  - Uses updated Global Monthly Retail Fuel Price Database, extending previous series through December 2020 and increasing coverage to 190 countries.
  - Adds product differentiation: premium and regular gasoline and diesel prices.
- CPI decomposition:
  - Analyzes the overall consumer price index (CPI) and 12 COICOP categories: food, transport, housing and energy, beverages, clothing, communication, education, health, equipment, recreation, hotels and other goods and services.
- Distributional reconstruction:
  - Uses multi-year household surveys to extract expenditure shares for the poorest quintile (the 20 percent poorest households) and the richest quintile (the 20 percent richest households).
  - Reconstructs quintile-specific CPI indexes as weighted averages of CPI components using these expenditure shares.
- Empirical contribution:
  - Provides empirical estimations of pass-through to quintile-specific inflation rates, complementing the literature that largely relies on partial and general equilibrium models.

### Channels, mechanisms, and heterogeneity drivers
- Direct channels:
  - Mechanical inclusion of fuel prices in CPI components consumed by households (transport, lighting, cooking).
- Indirect channels:
  - Cost pass-through via supply chains where fuel is an input to production and transport.
  - Producer responses: absorb costs (reduce margins, improve efficiency) or pass costs to consumers.
- Market and policy moderating factors:
  - Market structure (competition vs. monopoly) influences pass-through extent.
  - Government price regulation can weaken transmission of fuel price increases.
  - Labor market responses and monetary policy reactions (wage pressure, credibility of monetary policy) influence inflation persistence and amplitude.
- Cross-country heterogeneity:
  - Differences in energy intensity, labor market flexibility, central bank credibility, share of food and fuel in CPI, past inflation history, and governance quality drive variation in pass-through across countries.
  - Positive oil price shocks tend to have larger effects on inflation than negative shocks (asymmetric response).

### Policy implications and considerations highlighted
- Inflation management:
  - Acknowledging stronger and more persistent pass-through in developing countries is important for designing monetary and fiscal responses to fuel price shocks.
- Carbon pricing and equity:
  - Implementing carbon taxes (e.g., the $75 per ton benchmark) requires measures to mitigate adverse distributional impacts and inflationary consequences, particularly where current emissions prices (~$3 per ton) are far below the required level.
- Reform acceptability:
  - Mitigating measures such as targeted transfers or social safety nets are key to offsetting adverse impacts of fuel price increases or subsidy removals on vulnerable households and to ensure reform sustainability.
  - Evidence from country studies suggests partial subsidy removal plus targeted transfers can offset initial income decreases for the poor.
- Data policy:
  - High-frequency, disaggregated retail fuel price data are essential for accurate estimation and policy design; reliance on crude oil prices alone can lead to substantial underestimation of pass-through.

### Figures, annexes, and empirical outputs referenced (inventory from the source)
- Core figures include time series and impulse response analyses of gasoline and crude oil price effects on CPI and quintile-specific CPIs, and comparisons before/after tax.
- Annexes provide country samples, household expenditure share data by quintiles, and tables quantifying impacts of gasoline price changes on inflation for advanced and developing economies.

*Source: wpiea2021271-print-pdf - REFERENCES ____________________________________________________________22*

### 2016. In this paper, the dataset undergoes a major upgrade. The period of data availability now

### wpiea2021271-print-pdf - 2016. In this paper, the dataset undergoes a major upgrade. The period of data availability now

### Dataset upgrade and coverage
- Period of data availability now ranges from January 2000 to December 2020 for most countries (a few of them have historical data back to the 70s).
- Twenty-eight additional countries have been added to the dataset, bringing the total sample from 190 countries.
- Overall, the dataset consists of more than 425,000 country-month-product observations.
- Data are provided in local currencies and also converted in US dollars using the prevailing official exchange rates.
- The dataset now provides price data for premium and regular grades for gasoline and diesel to capture cases where different fuel grades are subject to different pricing policies.

### Observed regional patterns in retail gasoline prices
- Figure 2 (Average Gasoline Price by Region over Time, Jan 2010–Dec 2020) indicates:
  - Highest gasoline prices are observed in Europe.
  - South Asia, Middle East and North Africa, and North America exhibit the lower prices.
  - Regional differences exist in the response of retail gasoline prices to the sharp drop in crude oil prices amid the COVID-19 pandemic, with consumers in Europe, North and South America, and South Asia having benefited from the sharpest drop in gasoline prices in Q1 2020.

### Empirical data sources and sample used for regressions
- Monthly retail fuel prices: taken from the Global Monthly Retail Fuel Price Database described above.
- Monthly CPI and its 12 components: provided by the International Monetary Fund Macroeconomic and Financial Database, supplemented by data from national authorities.
- CPI components follow the COICOP classification and include: food, transport, housing and energy, beverages, clothing, communication, education, health, equipment, recreation, hotels and other goods and services.
- The overall CPI is the weighted average of its components, with weights being the average share of each component in total household expenditure.
- Regression sample: 122 countries (of which 48 advanced/high-income economies and 74 developing economies) with data during the period January 2000 to June 2019.

### Distributional analysis methodology (income quintiles)
- Objective: assess whether pass-through of changes in fuel prices to inflation has distributional implications by calculating CPI indices for different income groups using income-group-specific expenditure weights.
- Data challenge: lack of a centralized database on the share of CPI components in household expenditure for different income groups for a large sample of countries.
- Solution: use harmonized household survey databases from the World Bank.
  - For each country, households are grouped by income quintiles.
  - Calculate the average share of each category of goods and services in total expenditure, adjusted by household survey weights to ensure national representativeness.
- Application window for expenditure shares:
  - Expenditure shares from each household survey are applied to the CPI components over a 15-year window (4 years before the survey year and 10 year afterwards), unless data are available from another household survey during that period.
- Resulting quintile-level CPI series:
  - The monthly CPI index is calculated for each quintile in 84 countries in the sample during January 2000—June 2019, based on availability of household survey data.

*Source: https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021271-print-pdf.pdf*

### Annex Figure 1 shows the average household expenditure share across quintiles and CPI

### wpiea2021271-print-pdf - Annex Figure 1 shows the average household expenditure share across quintiles and CPI

### Household expenditure patterns and CPI component shares
- Significant heterogeneities across quintiles, CPI components and country income groups.
- Developing economies:
  - Food products dominate the household consumption basket.
  - Food share reaches 55 percent for the poorest quintile compared to 33 percent for the richest quintile.
  - Richest households devote on average a third of their expenditure to food products.
- Advancing (advanced) economies:
  - Housings account for the highest share in total household expenditure, followed by food products and transport services.
  - The poorest quintile in advanced economies spends 20 percent on food products.
- Cross-cutting patterns:
  - The food consumption share declines with household income in all country income groups.
  - Spending shares for transport, hospitality and recreation services increase with household income.

### Data coverage and construction of quintile CPIs
- Household survey data were available for 110 countries overall.
- To ensure cross-country comparability of quintile CPIs:
  - Required availability of expenditure shares for the 12 CPI components and the CPI indexes for the 12 components.
  - A minimum threshold of 80 percent was set: quintile CPIs computed only if data are available for at least 80 percent of goods and services of the household consumption basket.
  - This requirement reduced the sample to 84 countries.
- Note: Using a 10-year window centered at the year of the survey does not materially affect the findings but reduces the data sample.

### Model and methodology
- Estimation approach:
  - Local projection method (Jordà (2005)) augmented following Teulings and Zubanov (2014) to reduce potential bias.
  - Gasoline used as representative fuel product.
- Estimated model (summary):
  - Dependent variable: ∆Ln(CPI)_{i,t+h} (monthly CPI).
  - Key regressor: ∆Ln(RFP) (average retail gasoline price in local currency).
  - Includes lags, Teulings and Zubanov correction factor, time dummy τ, country-specific effects u_i, error term clustered at country-level.
  - The coefficient ϕ_h represents pass-through of gasoline prices to consumer price at horizon h.
  - Equation estimated sequentially by OLS for each horizon h up to maximum horizon H=11 (12 months).
- Application:
  - Same framework used to estimate distributional impact by replacing dependent variable with ratio of CPI of richest to poorest quintile.

### Pass-through of gasoline prices to consumer price inflation (key empirical results)
- Advanced economies:
  - A one percentage point increase in retail gasoline price leads to a 0.04 percentage point increase in consumer price inflation at peak level.
  - The pass-through dies out within 2 months after the gasoline price shock.
  - Peak pass-through in advanced economies is double that of developing economies.
- Developing economies:
  - A one percentage point increase in gasoline prices is estimated to push consumer prices up by 0.02 percentage point at peak.
  - Pass-through is more persistent: five to six months in developing economies, longer persistence than in advanced economies.
  - Cumulative impact on the CPI level is larger in developing countries because persistence more than offsets the lower peak pass-through.
- Robustness:
  - Results quantitatively comparable after controlling for (i) policy rate of the central bank, and (ii) exchange rate movements and international food prices.
- Proposed drivers:
  - Larger impact in developing countries: higher energy intensity and lower substitution effects.
  - Greater persistence: price and labor market rigidities and weak monetary policy credibility.

### Heterogeneity across CPI components
- Transport CPI:
  - Strongest response in both advanced and developing countries.
  - Magnitude larger in advanced economies, possibly due to regulated transport costs in developing economies.
- Housing and utilities:
  - Second largest response, with similar magnitude across advanced and developing countries.
  - Partly reflects direct inclusion of fuel price changes in this CPI component.
- Food prices:
  - React similarly to gasoline price shocks in both groups, but pass-through is more persistent in developing countries.
- Contribution to overall inflation:
  - Transport, housing and utilities, and food prices account for a combined 60 percent of household expenditure in advanced economies, and 70 percent in developing economies.
- Breadth of response:
  - Developing economies: 10 out of the 12 CPI components show a positive and significant response to an increase in gasoline prices.
  - Advanced economies: 7 out of the 12 CPI components show a positive and significant response.

### Crude oil prices vs retail gasoline prices
- Rationale:
  - Consumers consume refined fuel products, not crude oil; margins and government taxes matter.
  - Fuel taxes add substantially to retail fuel prices and vary across countries.
- EU example:
  - Fuel taxes account on average for about 60 percent of retail gasoline prices in June 2021.
  - By contrast, fuel taxes account for less than 20 percent in the US.
- Bias illustration (EU countries):
  - A 1 percentage point increase in crude oil prices is associated with a 0.015 percentage point increase in consumer price inflation at peak level.
  - A 1 percentage point increase in after-tax retail gasoline prices is associated with a 0.055 percentage point increase at peak.
  - Using before-tax retail gasoline price, the pass-through at peak level is 0.02 percentage point (close to crude oil price result).
  - Conclusion: using crude oil prices significantly understates the pass-through; underestimation is largely explained by not accounting for fuel tax dynamics.
  - Standardizing variables (mean zero, standard deviation one) yields the same qualitative finding, though magnitude of underestimation is smaller.

### Distributional impact of gasoline price shocks
- Metric:
  - Ratio of the CPI index of the richest quintile to the CPI index of the poorest quintile.
  - A positive and significant response of this ratio to gasoline price shocks denotes a progressive impact (CPI for poorest increases less than for richest).
  - A negative and significant response denotes a regressive impact.
- Advanced economies (Figure 8 results):
  - A 1 percent increase in gasoline prices raises the CPI of the richest households by about 0.02 percentage point more than the CPI of the poorest households in the first two months.
  - The marginal impact declines over time to zero from the 6th month onward (no differential impact after month 6).
  - Likely driver: higher expenditure shares on transport for richer households plus strong transport price response.
- Developing economies (Figure 9 results):
  - Similar larger impact on CPI of the richest households, but the differential impact declines only slightly and persists in positive territory for more than 12 months.
  - Contributing factors: higher transport cost and transport expenditure share, plus housings and food products.
  - Persistence likely reflects labor market rigidities and weaker monetary policy credibility.
- Quintile-specific pass-through (developing countries):
  - Peak pass-through of gasoline price shocks to inflation increases with household income (richest quintiles experience higher peak pass-through).
- Regional robustness:
  - Re-estimation for selected regions broadly confirms progressivity, which is stronger in Asia and Latin America than in Sub-Saharan Africa.

### Conclusions and policy implications
- Main empirical takeaways:
  - Peak pass-through in advanced economies is double that in developing economies (0.04 percentage point vs 0.02 percentage point per 1 percentage point gasoline price increase), but transmission is more protracted in developing economies, yielding a larger cumulative impact.
  - Using crude oil prices rather than retail fuel prices underestimates pass-through; fuel taxes are a key factor and should not be ignored.
  - Pass-through is more broad-based in developing countries (10 out of 12 components) than in advanced (7 out of 12).
  - Distributional impact of gasoline price shocks is typically progressive in both advanced and developing economies: richer households see larger CPI increases, but progressivity phases out within 6 months in advanced economies and persists beyond a year in developing countries.
- Policy recommendations:
  - Factor in knock-on inflationary and distributional effects when increasing fuel prices for pollution and environmental reasons.
  - Concurrent policies to mitigate inflationary impact:
    - Improve energy efficiency (e.g., technological innovations).
    - Boost availability of affordable and cleaner energy sources.
  - Reduce persistence of pass-through and risk of de-anchored inflation expectations by:
    - Improving labor market flexibility.
    - Strengthening monetary policy credibility.
    - Limiting price controls.
  - Fiscal and social policy:
    - Streamline fuel subsidies since they benefit the richest households more than the poorest.
    - Implement targeted mitigating measures to protect the poorest households, acknowledging that targeting takes time and effort but is achievable (experience from COVID-19 social safety net implementations cited).

*Source: Authors’ calculations and analysis provided in the content unit.*

### REFERENCES

### wpiea2021271-print-pdf - REFERENCES

### Major referenced topics (by theme)
- Oil price shocks, pass-through, and inflation dynamics
  - Key studies and datasets: Álvarez et al. 2011; Baumeister and Peersman 2013; Blanchard and Galí 2007; Castro and Jiménez-Rodríguez 2017; Chen 2009; Choi et al. 2018; Cologni and Manera 2008; De Gregorio et al. 2007; Gelos and Ustyugova 2017; Kilian 2009; Peersman and Robays 2012; Hammoudeh and Reboredo 2018.
- Fuel price pass-through to retail and distributional effects
  - Empirical and policy analyses: Abdallah and Kpodar 2020; Kpodar and Abdallah 2017; Caselli and Roitman 2015; Clements et al. 2003; Coady et al. 2006; Granado and Coady 2010; Kpodar and Djiofack 2010; Solie and Mu 2015; Siddig et al. 2014.
- Carbon pricing, poverty and distributional impacts
  - Cross-country / global comparative work and policy proposals: Dorband et al. 2019; Perry, Black, and Roaf 2021; Shang 2021.
- Methodology and macroeconomic transmission
  - Estimation and inference methods, fiscal multipliers, local projections: Jordà 2005; Jordà and Schularick and Taylor 2013; Ramey and Zubairy 2018; Auerbach and Gorodnichenko 2013; Auerbach and Gorodnichenko 2013 (two entries); Teulings and Zubanov 2014.
- Databases, classifications, and supporting analyses
  - Global retail fuel price database and household consumption classification: International Monetary Fund 2021; United Nations 2018; Bischof 2017 (graphing); Annex materials authored by the study team.

### Annex figures (captions and notes)
- Annex Figure 1. Average Household Expenditure Share by Quintiles and CPI Components (percent)
  - Sources: OECD, World Bank and Authors’ calculations
  - Notes: Data from the latest household survey available (see Annex 3).
- Annex Figure 2. Impulse Response Function of Consumer Price Inflation to Gasoline Price Changes in Advanced Economies – Additional Controls
  - Source: Authors’ calculations.
  - Notes: Additional controls include exchange rate, monetary policy rate and international food prices. Shaded area represents the 90 percent confidence interval.
- Annex Figure 3. Impulse Response Function of Consumer Price Inflation to Gasoline Price Changes in Developing Economies – Additional Controls
  - Source: Authors’ calculations.
  - Notes: Additional controls include exchange rate, monetary policy rate and international food prices. Shaded area represents the 90 percent confidence interval.
- Annex Figure 4. Impulse Response Function of Consumer Price Inflation to Changes in Crude Oil Prices, Gasoline Prices Before and After Tax – Standardized variables
  - Source: Authors’ calculations.
  - Notes: To estimate the impulse response function, the change in CPI, crude oil prices, and gasoline prices before and after tax have been rescaled to have a mean of zero and a standard deviation of one. Shaded area represents the 90 percent confidence interval. The sample consists of European Union countries.
- Annex Figure 5. Selected Developing Regions: Impulse Response Functions of the ratio of the Richest Quintile CPI to the Poorest Quintile CPI with respect to Gasoline Price Changes
  - Sub-Saharan Africa (32 countries)
    - Source: Authors’ calculations.
    - Notes: Shaded area represents the 90 percent confidence interval.
  - South and East Asia, and Pacific (6 countries)
    - Source: Authors’ calculations.
    - Notes: Shaded area represents the 90 percent confidence interval.
  - Latin America (4 countries)
    - Source: Authors’ calculations.
    - Notes: Shaded area represents the 90 percent confidence interval.

### Annex listings and datasets
- Annex 1. The Global Monthly Retail Fuel Price Database (2021): Country Sample
  - Includes countries and territories across regions: Europe & Central Asia; East Asia & Pacific; Latin America & Caribbean; Sub-Saharan Africa; Middle East & North Africa; South Asia; North America.
- Annex 2. Country Sample for the IRFs of Consumer Price Inflation to Gasoline Price Changes
  - Notes: Includes countries and territories. Categories include Advanced/High-income economies and Developing economies.
- Annex 3. Household Expenditure Share Data by Quintiles: Survey Years
  - Tabulated list of countries with Source and Survey years (sample entries shown in the source).

### Annex tables: fixed-effect estimations on gasoline price changes and inflation
- Annex Table 1. Impact of Gasoline Price Changes on Inflation in Advanced Economies
  - Notes. Fixed effect estimations. Robust standard errors in brackets. *,**,*** denote significance at 10 percent, 5 percent and 1 percent, respectively. The shaded line represents the coefficients of the impulse response function shown in Figure 3.
  - Selected panel statistics (as reported):
    - Observations: 8,3688,3208,2728,2248,1768,1288,080
    - Number of  countries: 48484848484848
    - Adjusted R2: 0.430.340.340.350.350.350.35
- Annex Table 2. Impact of Gasoline Price Changes on Inflation in Developing Economies
  - Notes. Fixed effect estimations. Robust standard errors in brackets. *,**,*** denote significance at 10 percent, 5 percent and 1 percent, respectively. The shaded line represents the coefficients of the impulse response function shown in Figure 4.
  - Selected panel statistics (as reported):
    - Observations: 9,1479,0859,0078,9298,8518,7738,695
    - Number of  countries: 74747474747474
    - Adjusted R2: 0.230.130.120.120.150.150.15

*References and annex captions reproduced from the source content.*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021271-print-pdf.pdf_
