## 1. Income and Price Elasticities of Energy Demand: Literature Review

## Source details

**Canonical URL:** [1. Income and Price Elasticities of Energy Demand: Literature Review](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13112.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13112.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13112.pdf.json)

---

### I. Introduction and motivation
- Motivation: analyze energy subsidies for Efficiency, Equity, Sustainability.
- Main aim: analyze a panel of cross-country data to explore the link between energy prices and energy consumption and estimate long-term own-price elasticity of energy demand.

### II. Empirical literature on elasticities — summary
- Short-term and long-term price elasticities reported in the literature span wide ranges:
  - Hamilton (2009): short-term nearly zero to -0.25; long-term -0.21 to -0.86.
  - IEA (1999): long-term price elasticities from -0.25 to -0.5 for different fuels and electricity demand.
  - IMF (April WEO 2011): short-term oil demand elasticity nearly zero; long-term -0.07.
- Income elasticities in literature broadly close to one in many studies:
  - Hamilton (2009): 0.88 to 1.32.
  - IMF (2011) suggests lower long-term income elasticities for oil consumption in some cases.

### III. Data, model, and framework
- Model (log-linear demand form):
  - ln(energy demand) = ln(α) + ln(γ_t) + ln(Θ_i) + δ ln(income_i,t) + β ln(price_i,t)
  - δ expected positive; β expected negative.
- Data sources and coverage:
  - Energy consumption (million tons of oil equivalent): British Petroleum 2011 Review, available 1965–2010 for 66 countries.
  - Population: United Nations.
  - GDP per capita (U.S. dollars) and U.S. CPI: IMF World Economic Outlook, 1965–2010.
  - Gasoline prices (proxy for energy prices): IMF Fiscal Affairs Department (2002–09) and GTZ (2010); dataset period 2002–10, setting time dimension to 2002–10.
  - Weather: U.S. NOAA processed into cold (average of 3 coldest months) and hot (average of 3 hottest months) indices; collapsed into time-invariant averages per country.
  - Advanced countries dummy: IMF WEO classification — sample: 29 advanced economies, 37 non-advanced economies.
- Final panel: highly balanced panel of 66 countries covering 2002–10 (with some missing gasoline price years for some countries).

### IV. Estimation strategy
- Primary estimator: OLS with robust standard errors adjusting for within-cluster correlation; time dummies included.
- Rationale: Short time dimension (nine years) makes OLS preferable for long-term elasticity estimation (reference to Baltagi and Griffin (1984)).
- Identification largely from cross-country variation; illustrative use of energy-share-in-GDP vs. gasoline price relation.

### V. Estimation results — main findings
- Good explanatory power: R-squared on the order of 0.75–0.81.
- Baseline estimates (Table 2, column (3)):
  - Income elasticity: around 0.77.
  - Long-term price elasticity (gasoline price proxy): around -0.3.
  - Weather effects:
    - winter 10°C colder → energy consumption +2.9 percent;
    - summer 10°C hotter → energy consumption +2.6 percent.
  - Advanced-countries dummy not statistically significant.
- Heterogeneity across country groups (Table 3):
  - Income elasticity lower in advanced countries: estimated advanced-country income elasticity ≈ 0.46 (interaction term negative and significant).
  - Price elasticity remains similar across groups (interaction insignificant).
  - Dropping two outlier countries with real price below $0.14 cents per liter (Table 3, column (2)) increases price elasticity magnitude to around -0.5 while interaction term for price remains insignificant.

### VI. Robustness checks
- Measurement error in gasoline prices (proxy for energy prices):
  - Replacing dependent variable with log oil consumption produces small increase in income elasticity and R-square but negligible change in price elasticity (Table 4, column (1)).
- Alternative estimators:
  - Between estimator results close to OLS (Table 4, column (2)), indicating cross-country variation is key.
  - Within estimator yields income and price elasticities not significantly different from zero (Table 4, column (3)), consistent with much smaller short-term elasticities in the literature.
- Time dummies:
  - Removing time dummies (Table 4, column (4)) yields similar income and price elasticities; income elasticity slightly lower and price elasticity slightly higher (in absolute value).
- Overall: results robust to measurement error, omitted variables concerns, and inclusion/exclusion of time dummies. Cross-country variation drives the estimated long-term elasticities.

### VII. Key quantitative results (exact figures)
- Estimated long-term price elasticity of energy demand: between -0.3 and -0.5.
- Estimated income elasticity of energy demand: in the order of 0.7–0.8 (baseline ~0.77; advanced-country estimate ~0.46).
- R-squared of baseline specifications: on the order of 0.75–0.81.
- Sample and period: 66 countries; period 2002–10; observations reported as 548 in baseline tables.
- Weather effects:
  - 10 degrees Celsius colder in winter → +2.9 percent energy consumption.
  - 10 degrees Celsius hotter in summer → +2.6 percent energy consumption.
- Outlier cutoff used in one robustness check: real price below $0.14 cents per liter excluded.

### VIII. Policy implications and recommendations
- Long-term perspective:
  - Given long-term price elasticity between -0.3 and -0.5, countries—especially those with large subsidies—can obtain significant gains from subsidy reform.
  - Removing subsidies saves areas A, B, and deadweight loss (dw) in standard supply-demand diagram; permanent net savings can be significant (example cited: Kuwait long-term net permanent saving approximately between 0.9–1.4 percent of GDP per year from Charap, Ribeiro da Silva, and Rodriguez (2012)).
- Short-term considerations:
  - Short-term price elasticity likely much lower than long-term elasticity → short-term consumer welfare losses are higher.
  - Policy prescriptions:
    - Accompany subsidy removal with safety nets targeted to lower-income segments.
    - Consider gradual approach to subsidy reform or more generous short-term safety nets to mitigate short-term welfare losses.
    - Governments choosing rebate schemes:
      - Rebating areas A and B would keep consumers indifferent and yield net savings equal to dw.
      - Rebating only area A yields net permanent savings equal to areas B and dw.
      - Targeted rebates to low-income consumers will yield savings depending on rebate scope.

### IX. Conclusions
- Main quantitative conclusions:
  - Long-term price elasticity of energy demand: between -0.3 and -0.5.
  - Income elasticity: ~0.7–0.8, with lower income elasticity for advanced countries (~0.46).
- Interpretation:
  - Estimates are in the middle to lower range of the empirical literature.
  - Price elasticity does not vary significantly across advanced vs. non-advanced countries; income elasticity is lower in advanced countries.
- Robustness and limitations:
  - Findings robust to measurement error, omitted-variable concerns, and inclusion/exclusion of time dummies.
  - Cross-country variation is main source of identifying variation; Within estimates suggest much smaller short-term elasticities.
  - Longer time-dimension datasets would be useful to further explore long-term vs. short-term elasticity differences.
- Policy takeaway:
  - Subsidy reform can yield significant long-term benefits, but short-term welfare costs necessitate careful sequencing and safety-net design (gradual reform or more generous short-term protection).

*Source: Excerpt from _wp13112 (IMF working paper).*

### 1. Income and Price Elasticities of Energy Demand: Literature Review ..................................4

### 1. Income and Price Elasticities of Energy Demand: Literature Review ..................................4

### Major themes and chapter structure
- The content unit is organized into four main sections:
  - "1. Income and Price Elasticities of Energy Demand: Literature Review" (page indicated as 4).
  - "2. Estimating Energy Demand—Baseline Estimations" (page indicated as 11).
  - "3. Estimating Energy Demand—Allowing for Different Income and Price Elasticities Across Country Groups" (page indicated as 12).
  - "4. Estimating Energy Demand—Robustness Checks" (page indicated as 15).

### Figures listed in the content unit
- Figure 1: "Energy Consumption and Gasoline Prices, 2010" (figure index 1).
- Figure 2: "Log of Share of Energy Consumption to (real) GDP and Log of Real Gasoline Prices in a Sample of 66 Countries, 2002–10" (figure index 2).
- Figure 3: "Cost and Deadweight Loss from Subsidies" (figure index 3).

### Intended analytical focus (as signaled by section titles)
- Literature review on income and price elasticities of energy demand.
- Baseline estimations of energy demand.
- Estimations that allow for heterogeneous income and price elasticities across country groups.
- Robustness checks of the estimations.

### Structural metadata preserved
- Section ordering and exact section titles preserved verbatim.
- Exact figure titles and sample period preserved verbatim: "2002–10" and "2010".
- Exact page references as presented: 4, 11, 12, 15.

*Source: _wp13112 - 1. Income and Price Elasticities of Energy Demand: Literature Review ..................................4*

### References .............................................................................................................

### _wp13112 - References .............................................................................................................

### I. Introduction
- Motivation for analyzing energy subsidies: Efficiency, Equity, Sustainability.
- Cited concerns and prior work include: G-20 Pittsburgh communiqué, 2009; Coady et al. (2010); IEA, OPEC, OECD, and World Bank (2010); Guillaume, Zytek, and Farzin (2011).
- Main aim: analyze a panel of cross-country data to explore the link between energy prices and energy consumption and estimate long-term own-price elasticity of energy demand.

### II. Empirical literature on responsiveness of energy consumption to prices
- Short-term and long-term price elasticities in literature span wide ranges:
  - Examples: Hamilton (2009): short-term nearly zero to -0.25; long-term -0.21 to -0.86.
  - IEA (1999): long-term price elasticities from -0.25 to -0.5 for different fuels and electricity demand.
  - IMF (April WEO 2011): short-term oil demand elasticity nearly zero; long-term -0.07.
- Income elasticities broadly close to one in many studies (Hamilton (2009): 0.88 to 1.32), though some studies (IMF (2011)) suggest lower long-term income elasticities for oil consumption.

### III. Reexamining (long-term) price elasticity — data and framework
- Analytical model (log-linear demand form):
  - ln(energy demand) = ln(α) + ln(γ_t) + ln(Θ_i) + δ ln(income_i,t) + β ln(price_i,t)
  - δ expected positive; β expected negative.
- Data sources and coverage:
  - Energy consumption (million tons of oil equivalent): British Petroleum 2011 Review, available 1965–2010 for 66 countries.
  - Population: United Nations.
  - GDP per capita (U.S. dollars) and U.S. CPI: IMF World Economic Outlook, 1965–2010.
  - Gasoline prices (proxy for energy prices): IMF Fiscal Affairs Department (2002–09) and GTZ (2010); dataset period 2002–10, setting time dimension to 2002–10.
  - Weather: U.S. NOAA processed into cold (average of 3 coldest months) and hot (average of 3 hottest months) indices; collapsed into time-invariant averages per country.
  - Advanced countries dummy: IMF WEO classification — sample: 29 advanced economies, 37 non-advanced economies.
- Final panel: highly balanced panel of 66 countries covering 2002–10 (with some missing gasoline price years for some countries).

### IV. Estimation strategy
- Primary estimator: OLS with robust standard errors adjusting for within-cluster correlation; time dummies included.
- Rationale: Short time dimension (nine years) makes OLS preferable for long-term elasticity estimation (reference to Baltagi and Griffin (1984)).
- Illustrative case: unit income elasticity implies using share of energy consumption in GDP vs. gasoline price; Figure 2 shows inverse relation.

### V. Estimation results — main findings
- Good explanatory power: R-squared on the order of 0.75–0.81.
- Baseline estimates (Table 2, column (3)):
  - Income elasticity: around 0.77.
  - Long-term price elasticity (gasoline price proxy): around -0.3.
  - Weather effects: winter 10°C colder → energy consumption +2.9 percent; summer 10°C hotter → energy consumption +2.6 percent.
  - Advanced-countries dummy not statistically significant.
- Heterogeneity analysis (Table 3):
  - Allowing different elasticities across advanced vs. non-advanced:
    - Income elasticity lower in advanced countries: estimated advanced-country income elasticity ≈ 0.46 (interaction term negative and significant).
    - Price elasticity remains similar across groups.
  - Dropping two outlier countries with real price below $0.14 cents per liter (Table 3, column (2)) increases price elasticity magnitude to around -0.5 while interaction term for price remains insignificant (price elasticity equal across groups).

### VI. Robustness checks
- Measurement error in gasoline prices (proxy for energy prices):
  - Replacing dependent variable with log oil consumption produces small increase in income elasticity and R-square but negligible change in price elasticity (Table 4, column (1)), suggesting limited bias from proxy measurement error.
- Alternative estimators:
  - Between estimator results close to OLS (Table 4, column (2)), indicating cross-country variation is key.
  - Within estimator yields income and price elasticities not significantly different from zero (Table 4, column (3)), consistent with much smaller short-term elasticities in the literature.
- Time dummies:
  - Removing time dummies (Table 4, column (4)) yields similar income and price elasticities; income elasticity slightly lower and price elasticity slightly higher (in absolute value).
- Overall: results robust to measurement error, omitted variables concerns related to advanced-country regulations, and inclusion/exclusion of time dummies. Cross-country variation drives the estimated long-term elasticities.

### VII. Key quantitative results (preserved exactly)
- Estimated long-term price elasticity of energy demand: between -0.3 and -0.5.
- Estimated income elasticity of energy demand: in the order of 0.7–0.8 (baseline ~0.77; advanced-country estimate ~0.46).
- R-squared of baseline specifications: on the order of 0.75–0.81.
- Sample and period: 66 countries; period 2002–10; observations reported as 548 in baseline tables.
- Weather effects: 10 degrees Celsius colder in winter → +2.9 percent energy consumption; 10 degrees Celsius hotter in summer → +2.6 percent energy consumption.
- Outlier cutoff: real price below $0.14 cents per liter excluded in one robustness check.

### VIII. Implications for subsidy reform and policy recommendations
- Long-term perspective:
  - Given long-term price elasticity between -0.3 and -0.5, countries—especially those with large subsidies—can obtain significant gains from subsidy reform.
  - Removing subsidies saves areas A, B, and deadweight loss (dw) in standard supply-demand diagram; permanent net savings can be significant (example cited: Kuwait long-term net permanent saving approximately between 0.9–1.4 percent of GDP per year from Charap, Ribeiro da Silva, and Rodriguez (2012)).
- Short-term considerations:
  - Short-term price elasticity likely much lower than long-term elasticity → short-term consumer welfare losses are higher.
  - Policy prescriptions:
    - Accompany subsidy removal with safety nets targeted to lower-income segments.
    - Consider gradual approach to subsidy reform or more generous short-term safety nets to mitigate short-term welfare losses.
    - Governments choosing rebate schemes:
      - Rebating areas A and B would keep consumers indifferent and yield net savings equal to dw.
      - Rebating only area A yields net permanent savings equal to areas B and dw.
      - Targeted rebates to low-income consumers will yield savings depending on rebate scope.

### IX. Conclusions
- Main quantitative conclusions:
  - Long-term price elasticity of energy demand: between -0.3 and -0.5.
  - Income elasticity: ~0.7–0.8, with lower income elasticity for advanced countries (~0.46).
- Interpretation:
  - Estimates are in the middle to lower range of the empirical literature.
  - Price elasticity does not vary significantly across advanced vs. non-advanced countries; income elasticity is lower in advanced countries.
- Robustness and limitations:
  - Findings robust to measurement error, omitted-variable concerns, and inclusion/exclusion of time dummies.
  - Cross-country variation is main source of identifying variation; Within estimates suggest much smaller short-term elasticities.
  - Longer time-dimension datasets would be useful to further explore long-term vs. short-term elasticity differences.
- Policy takeaway:
  - Subsidy reform can yield significant long-term benefits, but short-term welfare costs necessitate careful sequencing and safety-net design (gradual reform or more generous short-term protection).

*Italic: Source — Excerpt from _wp13112 (References section), IMF working paper content provided.*

### References

### References

### Energy subsidies and policy
- Charap, J., A. Ribeiro da Silva, and P. Rodriguez, (2012) “Fuel Subsidies and Energy Consumption—A Cross-Country Analysis” in Kuwait: Selected Issues and Statistical Appendix, Chapter 3 (Washington: International Monetary Fund).
- Coady, D., R. Gillingham, R. Ossowski, J. Piotrowski, S. Tareq, and J. Tyson (2010), “Petroleum Product Subsidies: Costly, Inequitable, and Rising,” IMF Staff Position Note, SPN/1005 (Washington: International Monetary Fund).
- IEA, OPEC, OECD, and World Bank (2010), “Analysis of the Scope of Energy Subsidies and Suggestions for the G-20 Initiative,” Prepared for Submission to the Toronto G-20 Summit Meeting (Toronto).
- IEA (1999), “World Energy Outlook—Looking at Energy Subsidies: Getting the Prices Right,” pp 71–87 (Paris: International Energy Agency).
- Parry, I. and K. Small (2005), “Does Britain or the United States Have the Right Gasoline Tax?” The American Economic Review, Vol. 95, (4), pp. 1276–1289.

### Oil prices, scarcity, and demand
- Hamilton, James (2009), “Understanding Crude Oil Prices,” Energy Journal, Vol. 30, (2), pp. 179–206.
- International Monetary Fund (2011), “Oil Scarcity, Growth, and Global Imbalances,” World Economic Outlook—April, Chapter 3 (Washington).
- Narayan, P and R. Smyth (2007), “A Panel Cointegration Analysis of the Demand for Oil in the Middle East,” Energy Policy, Vol. 35, pp. 6258–6265.

### Econometric and methodological references
- Baltagi, B. and J. Griffin (1984), “Short and Long Run Effects in Pooled Models,” International Economic Review, Vol. 5, (3), pp. 631–645.

*Source: _wp13112 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13112.pdf_