## Chasing the Sun and Catching the Wind: Energy Transition and Electricity Prices in Europe

## Source details

**Canonical URL:** [Chasing the Sun and Catching the Wind: Energy Transition and Electricity Prices in Europe](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022220-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022220-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022220-print-pdf.pdf.json)

---

### Abstract — Key findings
- Renewable energy is associated with a significant reduction in wholesale electricity prices in Europe: an average impact of 0.6 percent for each 1 percentage points increase in renewable share.
- Evidence for a nonlinear effect: higher the share of renewables, the greater its effect on wholesale electricity prices.
- Mixed evidence on volatility:
  - OLS shows no statistically significant relationship between renewable generation and volatility after controls.
  - Quantile regression yields mixed results but indicates a negative effect of renewables on volatility at the highest quantiles.
- Average share of renewables during the sample period: 14 percent.
- Policy implication summary: reforms can accelerate the green transition while minimizing volatility in electricity prices via grid modernization, increased interconnections, investment to reduce congestion, storage deployment, and diversification with nonhydrocarbon/non-weather-sensitive technologies.

### Context and motivation
- Europe experienced a record-breaking surge in energy prices driven by cyclical, structural, and geopolitical developments, including Russia's invasion of Ukraine.
- Wholesale electricity prices increased by more than 400 percent from an average of €35 per megawatt-hour (MWh) in 2020 to almost €250 per MWh in December 2021, and the average wholesale price rose above €500 per MWh in March 2022.
- As of 2021, fossil fuels accounted for about 42 percent of electricity; the share of solar and wind reached 16.4 percent of total electricity generation.
- Under the marginal pricing method, the most expensive technology needed to meet demand sets the final wholesale electricity price; recent spikes have been broadly driven by the cost of production at natural-gas power plants.

### Objective and contributions
- Twofold objective:
  - Present stylized facts on the energy matrix and electricity markets across Europe.
  - Explore the impact of solar and wind assets on the level and volatility of wholesale electricity prices.
- Methodological contribution:
  - Use a panel of monthly observations for 24 European countries during 2014–2021.
  - Employ alternative econometric methods including panel quantile regression with fixed effects to capture nonlinear effects and heterogeneity across price quantiles.

### Data, variables, and measurements
- Sample: balanced panel of monthly observations covering 24 European countries during 2014–2021.
- Main dependent variables:
  - Level of wholesale electricity prices (€/MWh), converted to monthly averages from daily ENTSO-E data.
  - Volatility of wholesale electricity prices measured as the standard deviation of daily prices within a month:
    - v(ep^m) = sqrt( (1/30) * sum_{d=1}^{30} (ep^d − ep^m)^2 )
- Main variable of interest: share of electricity generated by renewable sources, including solar and wind.
- Controls: total electricity load (GWh), monthly average crude oil import price (US$ per barrel), monthly average temperature (°C).
- Transformations: natural logarithm applied to level and volatility of wholesale prices and all exogenous series except temperature to improve stationarity and interpret coefficients as elasticities.

### Stylized facts and summary statistics
- Key sample statistics (selected):
  - Mean wholesale electricity price over the sample period: €58.6 per MWh.
  - Standard deviation of wholesale electricity prices: 75.4.
  - Cross-country variation: minimum €1.4 per MWh (Norway) and maximum €281 per MWh (Italy).
  - Distributional feature: sample kurtosis for level and volatility of wholesale electricity prices is markedly higher than 3, indicating fat tails.
  - Average renewables share in sample period: 14 percent.
- Detailed descriptive statistics (selected):
  - Wholesale electricity price (log-level dependent variable): Observations 3,459; Mean 58.675; Std. dev. 75.41; Skewness 0.21; Kurtosis 151.5; Minimum 1.4; Maximum 281.0.
  - Volatility of wholesale electricity price (log of standard deviation): Observations 3,458; Mean 21.3; Std. dev. 68.8; Skewness 17.13; Kurtosis 37.1; Minimum 0.3; Maximum 1,481.9.
  - Renewable energy (share in total electricity generation): Observations 3,459; Mean 1,827.6; Std. dev. 2,117.7; Skewness 2.915; Kurtosis 15.7; Minimum 0.02; Maximum 22,519.0.
  - Solar: Observations 2,902; Mean 689.61; Std. dev. 1,067.3; Skewness 1.84; Kurtosis 4.00; Minimum 0.07; Maximum 7,831.0.
  - Wind: Observations 3,459; Mean 1,142.3; Std. dev. 1,427.1; Skewness 5.14; Kurtosis 41.2; Minimum 0.02; Maximum 20,681.0.
  - Total electricity load: Observations 2,796; Mean 3,127,295; Std. dev. 7,732,635; Skewness 2.6; Kurtosis 6.0; Minimum 720.65; Maximum 52,223,569.
  - Crude oil import price: Observations 910; Mean 54.0; Std. dev. 12.8; Skewness -0.1; Kurtosis -0.1; Minimum 13.8; Maximum 87.4.
  - Temperature (monthly average): Observations 2,332; Mean 10.2 degrees °C; Std. dev. 7.9; Skewness 0.1; Kurtosis -0.7; Minimum -11.4 degrees °C; Maximum 26.9 degrees °C.
- Historical shares of electricity production by technology (percent of total):
  - Renewables accounted for 15.1 percent of total electricity generation in 2019, up from 0.02 percent in 1990.
  - Selected breakdown by technology:
    - 1990: Coal 38.0, Oil 9.6, Natural gas 8.6, Nuclear 27.3, Hydro 15.7, Solar 0.0, Wind 0.0
    - 2019: Coal 17.7, Oil 1.4, Natural gas 21.0, Nuclear 22.6, Hydro 15.9, Solar 3.8, Wind 11.3

### Empirical methodology
- Main specifications estimated with country fixed effects ηi and time fixed effects μt; robust standard errors clustered at the country level.
- Level equation:
  - ep_{i,t} = α RE_{i,t} + β X_{i,t} + η_i + μ_t + ε_{i,t}
  - ep_{i,t} = logarithm of wholesale electricity prices
  - RE_{i,t} = share of renewable electricity generation (total, solar, wind)
  - X_{i,t} = controls: log total electricity load, monthly average crude oil import price, monthly average temperature
- Volatility equation:
  - v(ep_{i,t}) = α RE_{i,t} + β X_{i,t} + η_i + μ_t + ε_{i,t}
  - v(ep_{i,t}) = logarithm of the standard deviation of wholesale electricity prices
- Estimation approaches:
  - OLS for baseline (country and time fixed effects; clustered robust SEs).
  - Panel quantile regression with fixed effects following Canay (2011); bootstrapped standard errors clustered by group to capture heterogeneity across quantiles (0.1 to 0.9).

### Empirical results — Level of prices
- OLS (level):
  - The share of renewables has a negative and statistically significant effect on the level of wholesale electricity prices at the 1 percent level after controls.
  - Interpretation: an increase of 1 percentage point in electricity produced by renewables is estimated to lower the level of wholesale electricity prices by about 0.6 percent on average.
  - Nonlinear effect: quadratic term indicates that higher shares of renewables have an increasing effect on price reduction.
  - Model fit example: Number of observations 486; Number of countries 24; Country FE Yes; Time FE Yes; Adj R^2 reported as 0.85 (level specifications).
- Panel quantile regression (level):
  - Dampening effect of renewables varies across quantiles:
    - Estimated effects reported range from -0.02 (including quadratic term) at the 10th quantile to -0.07 at the median and -1.40 at the 90th quantile.
  - Larger coefficients at higher quantiles imply renewables have a stronger effect in lowering prices at the extreme upper tail of the price distribution.

### Empirical results — Volatility of prices
- OLS (volatility):
  - Coefficient on the share of renewables is negative but not statistically significant when controlling for load, crude oil price, and temperature.
  - Coefficient on the quadratic term of renewables is positive—suggesting that higher shares of renewables may increase volatility (statistically insignificant in baseline).
- Quantile regression (volatility):
  - Mixed results overall.
  - Evidence that renewable energy can dampen price volatility at the highest quantiles.
  - Coefficients vary substantially across quantiles in sign, magnitude, and statistical significance.

### Technology-specific results
- Solar vs wind (selected findings):
  - Solar-based electricity generation:
    - More effective in lowering the level of wholesale electricity prices than wind power, after controlling for other factors.
    - Associated with a reduction in the volatility of wholesale electricity prices.
    - Quantile coefficients for solar (level) include values such as:
      - 0.10: -0.047 [0.059]; 0.20: -0.149*** [0.034]; 0.50: -0.322*** [0.075]; 0.90: 0.126 [0.308].
    - Quantile coefficients for solar (volatility) include values such as:
      - 0.10: -0.053 [0.066]; 0.20: -0.161*** [0.043]; 0.50: -0.348*** [0.084]; 0.90: 0.128 [0.315].
  - Wind-based electricity generation:
    - Has a positive—and statistically insignificant—effect on price volatility in OLS.
    - Quantile coefficients for wind (level) include values such as:
      - 0.10: -0.017 [0.030]; 0.50: -0.022 [0.057]; 0.90: 0.317 [0.223].
    - Quantile coefficients for wind (volatility) include values such as:
      - 0.10: 0.006 [0.030]; 0.30: 0.067* [0.033]; 0.90: 0.328 [0.228].
  - Granular results show substantial variation in magnitude and significance across quantiles and technologies.

### Interpretation and implications
- The green transition away from fossil fuels reduces the level of wholesale electricity prices on average.
- Volatility implications are mixed and depend on the share and mix of renewable technologies as well as the position in the price distribution:
  - Renewables tend to lower prices more at the high end of the price distribution.
  - Intermittent renewable penetration may increase volatility in some contexts, but solar generation appears to reduce volatility while wind does not.
- Findings are broadly consistent with prior literature that reports mixed results on volatility effects of renewables.

### Policy recommendations and implications
- To maximize the price-dampening effect of renewables while limiting volatility, pursue:
  - Modernization and closer integration of electricity grids across Europe.
  - Increased investment to reduce congestion on transmission lines.
  - Deployment of volatility-dampening technologies such as storage.
  - Better electricity interconnections to increase cross-country trade and efficient allocation given intermittent production.
  - Consider a comprehensive continental electricity market under a multilateral coordination mechanism to harmonize policies, legal/regulatory frameworks, identify strategic projects, and secure investments.
  - Prefer targeted measures such as income support to shield consumers during price shocks rather than wholesale market intervention.
  - Achieve a better balance between solar and wind and diversify the matrix with nonhydrocarbon and non-weather-sensitive technologies (e.g., next-generation geothermal and nuclear) to provide stable, consistent electricity streams.

### Quantitative policy scenario examples (based on empirical estimates)
- Increasing renewables from average 14 percent to 30 percent would lower wholesale electricity prices by 8.8 percent.
- Increasing renewables to 50 percent would lower wholesale electricity prices by almost 20 percent.

*Source: Serhan Cevik and Keitaro Ninomiya, "Chasing the Sun and Catching the Wind: Energy Transition and Electricity Prices in Europe", IMF Working Paper WP/22/220 — Sections 1–3.*

### Section 1

### Chasing the Sun and Catching the Wind: Energy Transition and Electricity Prices in Europe

### Abstract — Key findings
- Renewable energy is associated with a significant reduction in wholesale electricity prices in Europe: an average impact of 0.6 percent for each 1 percentage points increase in renewable share.
- Evidence for a nonlinear effect: higher the share of renewables, the greater its effect on wholesale electricity prices.
- Mixed evidence on volatility: OLS shows no statistically significant relationship between renewable generation and volatility after controls; quantile regression yields mixed results but indicates a negative effect of renewables on volatility at the highest quantiles.
- Average share of renewables during the sample period: 14 percent.
- Policy implication summary: reforms can accelerate the green transition while minimizing volatility in electricity prices via grid modernization, increased interconnections, investment to reduce congestion, storage deployment, and diversification with nonhydrocarbon/non-weather-sensitive technologies.

### Context and motivation
- Europe experienced a record-breaking surge in energy prices driven by cyclical, structural, and geopolitical developments, including Russia's invasion of Ukraine.
- Wholesale electricity prices increased by more than 400 percent from an average of €35 per megawatt-hour (MWh) in 2020 to almost €250 per MWh in December 2021, and the average wholesale price rose above €500 per MWh in March 2022.
- As of 2021, fossil fuels accounted for about 42 percent of electricity; the share of solar and wind reached 16.4 percent of total electricity generation.
- Under the marginal pricing method, the most expensive technology needed to meet demand sets the final wholesale electricity price; recent spikes have been broadly driven by the cost of production at natural-gas power plants.

### Objective and contributions
- Twofold objective:
  - Present stylized facts on the energy matrix and electricity markets across Europe.
  - Explore the impact of solar and wind assets on the level and volatility of wholesale electricity prices.
- Methodological contribution:
  - Use a panel of monthly observations for 24 European countries during 2014–2021.
  - Employ alternative econometric methods including panel quantile regression with fixed effects to capture nonlinear effects and heterogeneity across price quantiles.

### Data, variables, and measurements
- Sample: balanced panel of monthly observations covering 24 European countries during 2014–2021.
- Main dependent variables:
  - Level of wholesale electricity prices (€/MWh), converted to monthly averages from daily ENTSO-E data.
  - Volatility of wholesale electricity prices measured as the standard deviation of daily prices within a month:
    - v(ep^m) = sqrt( (1/30) * sum_{d=1}^{30} (ep^d − ep^m)^2 )
- Main variable of interest: share of electricity generated by renewable sources, including solar and wind.
- Controls: total electricity load (GWh), monthly average crude oil import price (US$ per barrel), monthly average temperature (°C).
- Transformations: natural logarithm applied to level and volatility of wholesale prices and all exogenous series except temperature to improve stationarity and interpret coefficients as elasticities.

### Stylized facts and summary statistics
- Mean wholesale electricity price over the sample period: €58.6 per MWh.
- Standard deviation of wholesale electricity prices: 75.4.
- Cross-country variation: minimum €1.4 per MWh (Norway) and maximum €281 per MWh (Italy).
- Distributional feature: sample kurtosis for level and volatility of wholesale electricity prices is markedly higher than 3, indicating fat tails.
- Average renewables share in sample period: 14 percent.

### Empirical results — Level of prices
- OLS and panel quantile regression results:
  - Renewable-based energy lowers the average level of wholesale electricity prices; estimated elasticity: −0.6 percent per 1 percentage points increase in renewable share.
  - Nonlinear effect: coefficients indicate larger dampening effects on prices at higher renewable penetration.
  - Quantile regression shows negative effect across quantiles with varying magnitudes; splitting renewables into solar and wind yields more granular insights.

### Empirical results — Volatility of prices
- OLS analysis:
  - Coefficient on the share of renewables is negative but not statistically significant when controlling for load, crude oil price, and temperature.
  - Coefficient on the quadratic term of renewables is positive—suggesting that higher shares of renewables may increase volatility.
- Quantile regression:
  - Mixed results overall.
  - Evidence that renewable energy can dampen price volatility at the highest quantiles.
- Interpretation:
  - Intermittent and volatile production from renewables can contribute to instability in electricity provision and potentially higher prices through risk premia demanded by distribution companies.

### Policy recommendations and implications
- To maximize the price-dampening effect of renewables while limiting volatility, pursue:
  - Modernization and closer integration of electricity grids across Europe.
  - Increased investment to reduce congestion on transmission lines.
  - Deployment of volatility-dampening technologies such as storage.
  - Better electricity interconnections to increase cross-country trade and efficient allocation given intermittent production.
  - Consider a comprehensive continental electricity market under a multilateral coordination mechanism to harmonize policies, legal/regulatory frameworks, identify strategic projects, and secure investments.
  - Prefer targeted measures such as income support to shield consumers during price shocks rather than wholesale market intervention.
  - Achieve a better balance between solar and wind and diversify the matrix with nonhydrocarbon and non-weather-sensitive technologies (e.g., next-generation geothermal and nuclear) to provide stable, consistent electricity streams.

### Quantitative policy scenario examples (based on empirical estimates)
- Increasing renewables from average 14 percent to 30 percent would lower wholesale electricity prices by 8.8 percent.
- Increasing renewables to 50 percent would lower wholesale electricity prices by almost 20 percent.

*Source: Serhan Cevik and Keitaro Ninomiya, "Chasing the Sun and Catching the Wind: Energy Transition and Electricity Prices in Europe", IMF Working Paper WP/22/220 — Section 1.*

### Section 2

### Section 2

### Data and descriptive statistics
- Sample: panel of 24 countries in Europe over the period 2014–2021.
- Wholesale electricity price (log-level dependent variable):
  - Observations: 3,459
  - Mean: 58.675
  - Std. dev.: 75.41
  - Skewness: 0.21
  - Kurtosis: 151.5
  - Minimum: 1.4
  - Maximum: 281.0
- Volatility of wholesale electricity price (log of standard deviation):
  - Observations: 3,458
  - Mean: 21.3
  - Std. dev.: 68.8
  - Skewness: 17.13
  - Kurtosis: 37.1
  - Minimum: 0.3
  - Maximum: 1,481.9
- Renewable energy (share in total electricity generation):
  - Observations: 3,459
  - Mean: 1,827.6
  - Std. dev.: 2,117.7
  - Skewness: 2.915
  - Kurtosis: 15.7
  - Minimum: 0.02
  - Maximum: 22,519.0
- Solar:
  - Observations: 2,902
  - Mean: 689.61
  - Std. dev.: 1,067.3
  - Skewness: 1.84
  - Kurtosis: 4.00
  - Minimum: 0.07
  - Maximum: 7,831.0
- Wind:
  - Observations: 3,459
  - Mean: 1,142.3
  - Std. dev.: 1,427.1
  - Skewness: 5.14
  - Kurtosis: 41.2
  - Minimum: 0.02
  - Maximum: 20,681.0
- Total electricity load:
  - Observations: 2,796
  - Mean: 3,127,295
  - Std. dev.: 7,732,635
  - Skewness: 2.6
  - Kurtosis: 6.0
  - Minimum: 720.65
  - Maximum: 52,223,569
- Crude oil import price:
  - Observations: 910
  - Mean: 54.0
  - Std. dev.: 12.8
  - Skewness: -0.1
  - Kurtosis: -0.1
  - Minimum: 13.8
  - Maximum: 87.4
- Temperature (monthly average):
  - Observations: 2,332
  - Mean: 10.2 degrees °C
  - Std. dev.: 7.9
  - Skewness: 0.1
  - Kurtosis: -0.7
  - Minimum: -11.4 degrees °C
  - Maximum: 26.9 degrees °C
- Historical shares of electricity production by technology (percent of total):
  - Renewables accounted for 15.1 percent of total electricity generation in 2019, up from 0.02 percent in 1990.
  - The breakdown by technology (selected years):
    - 1990: Coal 38.0, Oil 9.6, Natural gas 8.6, Nuclear 27.3, Hydro 15.7, Solar 0.0, Wind 0.0
    - 2019: Coal 17.7, Oil 1.4, Natural gas 21.0, Nuclear 22.6, Hydro 15.9, Solar 3.8, Wind 11.3

### Seasonality and stationarity
- Electricity prices show significant seasonality and positive skewness (greater probability of large increases in price and volatility than large declines).
- Seasonality is controlled for by including total electricity load and temperature.
- Panel unit root tests (Im-Pesaran-Shin (2003)) indicate variables are stationary after logarithmic transformation; analysis proceeds in log-levels.

### Empirical methodology
- Two main specifications estimated (country fixed effects ηi and time fixed effects μt; robust standard errors clustered at the country level):
  - Level equation:
    - ep_{i,t} = α RE_{i,t} + β X_{i,t} + η_i + μ_t + ε_{i,t}
    - ep_{i,t} = logarithm of wholesale electricity prices
    - RE_{i,t} = share of renewable electricity generation (total, solar, wind)
    - X_{i,t} = controls: log total electricity load, monthly average crude oil import price, monthly average temperature
  - Volatility equation:
    - v(ep_{i,t}) = α RE_{i,t} + β X_{i,t} + η_i + μ_t + ε_{i,t}
    - v(ep_{i,t}) = logarithm of the standard deviation of wholesale electricity prices
- Estimation approaches:
  - OLS for baseline (country and time fixed effects; clustered robust SEs).
  - Panel quantile regression with fixed effects following Canay (2011); bootstrapped standard errors clustered by group (Angrist and Pischke (2009)) to capture heterogeneity across the conditional distribution (quantiles from 0.1 to 0.9).

### Main empirical findings
- OLS (level of prices):
  - The share of renewables has a negative and statistically significant effect on the level of wholesale electricity prices at the 1 percent level after controls.
  - Interpretation: an increase of 1 percentage point in electricity produced by renewables is estimated to lower the level of wholesale electricity prices by about 0.6 percent on average.
  - Nonlinear effect: quadratic term indicates that higher shares of renewables have an increasing effect on price reduction (the total impact of an additional 1 percentage point would increase to 0.6 percent on average).
  - Illustrative calculations reported:
    - Raising the share of renewables from an average 14 percent to 30 percent would lower wholesale electricity prices by 8.8 percent.
    - If the share of solar and wind reaches 50 percent, prices would lower by almost 20 percent.
  - Model fit example: Number of observations 486; Number of countries 24; Country FE Yes; Time FE Yes; Adj R^2 reported as 0.85 (level specifications).
- OLS (volatility of prices):
  - Mixed results: no statistically significant relationship between renewable share and price volatility at conventional levels in the baseline OLS.
  - Coefficient on renewable share is negative while coefficient on its quadratic term is positive (indicative that higher intermittent renewable share may increase volatility), but statistically insignificant.
- Quantile regression (level of prices):
  - The dampening effect of renewables varies across quantiles:
    - Estimated effects range from -0.02 (including quadratic term) at the 10th quantile to -0.07 at the median and -1.40 at the 90th quantile.
  - Larger coefficients at higher quantiles imply renewables have a stronger effect in lowering prices at the extreme upper tail of the price distribution (countries/times with relatively higher wholesale prices).
- Quantile regression (volatility of prices):
  - Coefficients vary substantially across quantiles in sign, magnitude, and statistical significance.
  - Renewable generation appears to have a stronger dampening effect on price volatility at the highest quantiles than at the median and lowest quantiles.

### Technology-specific results
- Disaggregating renewables into solar and wind (results reported in appendix tables):
  - Solar-based electricity generation:
    - More effective in lowering the level of wholesale electricity prices than wind power, after controlling for other factors.
    - Associated with a reduction in the volatility of wholesale electricity prices.
  - Wind-based electricity generation:
    - Has a positive—and statistically insignificant—effect on price volatility in OLS.
  - Quantile analysis confirms:
    - Solar power tends to have a stronger dampening effect on both the level and volatility of wholesale electricity prices than wind-based generation.
  - Granular results show substantial variation in magnitude and significance across quantiles and technologies.

### Interpretation and implications
- The green transition away from fossil fuels reduces the level of wholesale electricity prices on average.
- Volatility implications are mixed and depend on the share and mix of renewable technologies as well as the position in the price distribution:
  - Renewables tend to lower prices more at the high end of the price distribution.
  - Intermittent renewable penetration may increase volatility in some contexts, but solar generation appears to reduce volatility while wind does not.
- Findings are broadly consistent with prior literature that reports mixed results on volatility effects of renewables.

*wpiea2022220-print-pdf - Section 2*

### Section 3

### Section 3

### Methodology
- Monthly data on 24 European power markets during the period 2014–2021.
- Econometric methods include a panel quantile regression approach to analyze the level and volatility of wholesale electricity prices and capture nonlinearity effects of the green transition on various price quantiles.

### Key empirical findings on price levels
- Renewable-based energy lowers the average level of wholesale electricity prices.
- The estimated coefficient on the share of renewables in total electricity generation implies that an increase of 1 percentage points in electricity produced by renewables lowers wholesale electricity prices by 0.6 percent on average.
- Evidence of a nonlinear effect: higher the share of renewables, the greater its effect on wholesale electricity prices.
- Quantile regression estimates indicate the share of renewables has a negative effect on price levels but at different magnitudes across quantiles, with more granular differences when splitting renewables into solar and wind.
- The price-dampening effect of renewables is economically significant and highly likely to be underestimated because of the relatively low share of renewables in the energy matrix—an average of 14 percent during the sample period.

### Quantile-estimation highlights (solar and wind on price levels)
- Solar (level of wholesale electricity prices), quantile coefficients (bootstrapped standard errors in brackets):
  - 0.10: -0.047 [0.059]
  - 0.20: -0.149*** [0.034]
  - 0.30: -0.200** [0.070]
  - 0.40: -0.244*** [0.069]
  - 0.50: -0.322*** [0.075]
  - 0.60: -0.345* [0.149]
  - 0.70: -0.396 [0.266]
  - 0.80: -0.690 [0.386]
  - 0.90: 0.126 [0.308]
- Solar^2 (level), quantile coefficients:
  - 0.10: -0.006 [0.005]
  - 0.20: -0.014*** [0.003]
  - 0.30: -0.018** [0.006]
  - 0.40: -0.023*** [0.006]
  - 0.50: -0.030*** [0.007]
  - 0.60: -0.032 [0.017]
  - 0.70: -0.032 [0.027]
  - 0.80: -0.045 [0.032]
  - 0.90: -0.001 [0.027]
- Wind (level), quantile coefficients:
  - 0.10: -0.017 [0.030]
  - 0.20: 0.007 [0.028]
  - 0.30: 0.041 [0.030]
  - 0.40: 0.033 [0.051]
  - 0.50: -0.022 [0.057]
  - 0.60: -0.091 [0.667]
  - 0.70: -0.232 [0.902]
  - 0.80: -0.339 [0.220]
  - 0.90: 0.317 [0.223]
- Wind^2 (level), quantile coefficients show mostly small magnitudes (e.g., 0.001–0.014) with reported standard errors.

### Findings on price volatility
- Mixed results regarding the impact of intermittent renewable generation on the volatility of wholesale electricity prices.
- Standard analysis shows no statistically significant relationship between renewable electricity generation and price volatility in Europe during the sample period after controlling for total electricity load, crude oil import prices, and average temperature.
- The coefficient on the share of renewables is negative while the coefficient on its quadratic term is positive—indicating that higher the share of renewables, the greater its effect on the volatility of wholesale electricity prices.
- Quantile regression results for volatility are mixed, but there is some evidence that renewable energy has a dampening effect on price volatility at the highest quantiles.
- The intermittent and volatile nature of renewable generation remains a concern because it may contribute to higher prices through greater instability and higher risk premia borne by distribution companies.

### Quantile-estimation highlights (solar and wind on price volatility)
- Solar (volatility of wholesale electricity prices), quantile coefficients:
  - 0.10: -0.053 [0.066]
  - 0.20: -0.161*** [0.043]
  - 0.30: -0.212** [0.077]
  - 0.40: -0.274*** [0.072]
  - 0.50: -0.348*** [0.084]
  - 0.60: -0.368* [0.172]
  - 0.70: -0.417 [0.274]
  - 0.80: -0.714 [0.396]
  - 0.90: 0.128 [0.315]
- Solar^2 (volatility), quantile coefficients mirror level results with small negative coefficients at several quantiles (e.g., -0.007 to -0.046).
- Wind (volatility), quantile coefficients:
  - 0.10: 0.006 [0.030]
  - 0.20: 0.013 [0.031]
  - 0.30: 0.067* [0.033]
  - 0.40: 0.057 [0.057]
  - 0.50: 0.000 [0.059]
  - 0.60: -0.089 [0.713]
  - 0.70: -0.221 [0.934]
  - 0.80: -0.330 [0.229]
  - 0.90: 0.328 [0.228]
- Wind^2 (volatility), quantile coefficients typically small (e.g., 0.001–0.014) with reported standard errors.

### Quantitative examples and sample statistics cited
- Average share of renewables during the sample period: 14 percent.
- Policy scenario projections from the empirical analysis:
  - Increasing the share of renewables in electricity production in Europe from an average level of 14 percent during the period 2014–2021 to 30 percent would lower wholesale electricity prices by 8.8 percent.
  - If the share of solar and wind reaches 50 percent, wholesale electricity prices would be lower by almost 20 percent.
- Sample sizes and panel details in appendix tables:
  - Number of countries: 24.
  - Number of observations reported in tables: 486, 465, 323 (varies by specification).
  - Adjusted R2 values reported in Appendix Table A1 include values such as 0.85, 0.84, 0.90, 0.91, 0.76, 0.84, 0.85 (by specification).

### Policy implications and recommendations
- The transition to low-carbon power generation is necessary for mitigating socioeconomic consequences of climate change and strengthening energy security in Europe.
- To maximize the price-dampening effect of green power while minimizing volatility, policymakers should:
  - Pursue reforms for modernization and closer integration of electricity grids throughout Europe.
  - Increase investment to reduce congestion on transmission lines.
  - Introduce volatility-dampening technology solutions like storage.
  - Build better electricity interconnections to increase cross-country electricity trade and improve efficient allocation of electricity given intermittent renewable production.
  - Consider a comprehensive continental electricity market under supervision of a multilateral mechanism for cross-country coordination with a mandate to harmonize energy policies and legal regulatory frameworks, identify strategic projects, and secure necessary investments.
  - In the current environment, to mitigate the impact of high energy prices, prioritize targeted measures such as income support over intervention in the wholesale market.
  - Achieve a better balance between solar and wind power and diversify the energy matrix with nonhydrocarbon and non-weather sensitive technologies—examples cited include a new generation of geothermal and nuclear power plants to provide a stable and consistent stream of electricity across Europe.

*Source: wpiea2022220-print-pdf - Section 3.*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022220-print-pdf.pdf_