## wpiea2023120-print-pdf - Section V offers concluding remarks with policy implications

## Source details

**Canonical URL:** [wpiea2023120-print-pdf - Section V offers concluding remarks with policy implications](https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023120-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2023/english/wpiea2023120-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2023/english/wpiea2023120-print-pdf.pdf.json)

---

### Data and sample
- Panel dataset of annual observations covering 25 countries over the period 1970–2020 drawn from the OECD.
- Countries in the sample include: Australia, Austria, Belgium, Canada, Czech Republic, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, Korea, Luxembourg, the Netherlands, New Zealand, Norway, Portugal, the Slovak Republic, Spain, Sweden, Switzerland, the United Kingdom, and the United States.
- Dependent variables:
  - Environmental outcomes: (i) CO2 emissions in metric tons per capita, (ii) GHG emissions in metric tons per capita, (iii) GHG emissions per unit of GDP.
  - Green growth indicators: (i) share of renewable energy supply, (ii) number of environment-related inventions per capita, (iii) number of patents for environment-related technologies per capita.
- Main explanatory variables:
  - Structural reforms in the energy sector (electricity and gas) from a narrative database of major policy changes in product market regulation (Duval et al. (2018) updated by Wiese et al. (2023) until 2020).
  - Reform coding: 0 in non-reform years, 1 in reform years (liberalizing/deregulating), and -1 in counter-reform years (tightening/regulating).
- Table 1 summary (1970-2020):
  - Product market reforms: Number of reforms 235; Number of counter-reforms 4; Reforms (% of total) 98.3; Counter-reforms (% of total) 1.7.
  - Electricity sector: Number of reforms 48; Number of counter-reforms 0; Reforms (% of total) 100.0; Counter-reforms (% of total) 0.0.
  - Gas sector: Number of reforms 49; Number of counter-reforms 0; Reforms (% of total) 100.0; Counter-reforms (% of total) 0.0.
- Environmental Policy Stringency (EPS) Index:
  - OECD EPS index scaled from 0 (not stringent at all) to 6 (very stringent); covers 28 OECD countries and 6 emerging market economies over 1990–2020.
  - EPS used to test state-dependent impacts of energy reforms.

### Econometric methodology
- Estimation approach:
  - Local Projections (LP) method to estimate impulse response functions (IRFs) of environmental outcomes and green growth to structural reforms in electricity and gas.
  - Rationale: LP is flexible for panel data with fixed effects, economizes parameters relative to VARs, and is well suited to nonlinearities and state-dependent analysis.
- Baseline specification (cumulative responses over a five-year horizon):
  - Equation (1): y_{t+k,i} − y_{t-1,i} = α_i + τ_t + β_k SR_{i,t} + θ X_{i,t} + ε_{i,t}
    - y denotes the environmental or green-growth proxy.
    - α_i and τ_t are country and time fixed effects.
    - β_k denotes the cumulative response k years after reform; SR_{i,t} denotes structural reforms in electricity and gas.
    - Control variables: 2 lags of the structural reform shock, 2 lags of real GDP growth, inflation, and the dependent variable (lag length chosen by Akaike’s information criterion).
  - Estimation via Ordinary Least Squares (OLS).
  - Standard errors: Spatial Correlation Consistent (SCC) Driscoll-Kraay (1998) standard errors; also report 90 percent and 68 percent confidence bands.
- State-dependent (nonlinear) specification:
  - Equation (2): y_{t+k,i} − y_{t-1,i} = α_i + τ_t + β_k^L F(z_{i,t}) SR_{i,t} + β_k^H (1−F(z_{i,t})) SR_{i,t} + θ X_{i,t} + ε_{i,t}
  - Weighting function F(z) = exp(γ z) / (1 + exp(γ z)), γ>0; z is the EPS index normalized to zero mean and unit variance.
  - Interpretation: β_k^L captures impact in low-EPS states (F(z)≈1 as z→−∞); β_k^H captures impact in high-EPS states (1−F(z)≈1 as z→+∞).
  - Parameter choice: γ = 1.5 (results robust for γ between 1 and 4).
- Robustness and identification:
  - Use of Driscoll-Kraay (1998) SCC standard errors to handle heteroskedasticity, autocorrelation, and cross-country correlations.
  - Robustness checks include excluding country fixed effects, adding country-specific time trends, controlling for reforms in additional policy areas, and instrumental variable (IV) estimation.
  - IV strategy: political-economy instruments from the World Bank Database of Political Institutions:
    - Parties (discrete: left=3, center=2, right=1), System (discrete: parliamentary=2, assembly-elected=1, presidential=0), Fragmentation (continuous 0–1), Democ (Polity IV index normalized 0–1).
  - Instrument validity assessed using Kleibergen-Paap and Hansen statistics; reported tests generally reject underidentification and do not reject instrument validity.

### Main empirical findings
- Baseline (unconditional) results (cumulative five-year horizon):
  - Environmental outcomes:
    - Structural reforms in electricity and gas markets lead to higher CO2 and GHG emissions per capita, but these unconditional increases are statistically insignificant and surrounded by great uncertainty.
    - GHG emissions per unit of GDP declines below the initial level over the five-year period, showing signs of decoupling between economic growth and emissions.
  - Green growth:
    - Structural reforms lower the supply of renewable energy as a share of total energy supply initially; this negative effect is not persistent over the long run.
    - Structural reforms stimulate a sustained increase in the number of environment-related inventions and patents per capita beyond the initial level.
    - Note: energy-sector privatization in baseline leads to higher emissions across all measures, including GHG emissions per unit of GDP.
- Heterogeneity by reform type:
  - Similar response patterns found for “market access and structure” reforms and “public ownership” (privatization) reforms; IRFs for electricity and gas reforms separately are consistent with baseline results.
- State-dependent results (role of initial environmental policy stringency):
  - Striking differences by EPS:
    - In countries with stronger environmental regulations and pro-climate policies (high EPS), structural reforms in the energy sector deliver a significant and persistent decline in CO2 and GHG emissions per capita and in GHG emissions per unit of GDP in the first year and over the long run.
    - In low-EPS countries, emissions continue to increase following energy-sector reforms.
    - Reforms increase the share of renewable energy and development of environmental technologies in high-EPS countries, while the impact is the opposite in low-EPS countries.
  - Energy-sector privatization results in significantly better environmental outcomes and green growth in countries with stronger environmental regulations.
- Robustness exercises:
  - Excluding country fixed effects suggests the bias from the interaction of fixed effects and country-specific developments is negligible.
  - Including country-specific time trends leaves qualitative findings unchanged.
  - Controlling for reforms in additional areas yields results consistent with the baseline.
  - IV estimates using political-economy instruments broadly confirm baseline findings, suggesting endogeneity is not a serious concern.

### Key empirical takeaways reiterated
- Electricity and gas sector reforms so far have failed in bringing about a reduction in CO2 and GHG emissions per capita.
- There is some evidence that electricity and gas sector reforms are more effective in lowering GHG emissions per unit of GDP.
- Electricity and gas sector reforms are not associated with higher supply of renewable energy as a share of total energy supply.
- Electricity and gas sector reforms appear to stimulate a sustained increase in the number of environmental inventions and patents per capita over the medium term.
- Strong evidence of nonlinear effects: market-oriented electricity and gas sector reforms lead to better environmental outcomes and green growth in countries with stronger environmental regulations.

### Policy implications and recommendations (Section V)
- Climate challenge and required mitigation:
  - GHG emissions are projected to increase by only 7.5 percent by 2030 compared to the required reduction of 55 percent just to keep global warming below 1.5°C.
- Policy message:
  - Achieving necessary emissions reductions and promoting green growth requires a new development model that better balances income growth and environmental priorities by:
    - Modernizing the energy matrix away from fossil fuels.
    - Increasing efficiency in the distribution and use of energy.
    - Implementing structural reforms and policies designed to alter behavior throughout the economy.
- Role of energy-sector structural reforms:
  - Structural reforms in electricity and gas can contribute to climate change mitigation and promote green growth, but their environmental effectiveness depends critically on the strength of concurrent environmental policies (EPS).
  - In countries with strong environmental policy frameworks (high EPS), energy-sector reforms are more likely to lower emissions and foster renewable energy expansion and clean-technology innovation.
  - In countries with weak environmental policies (low EPS), similar reforms may lead to higher emissions and reduced renewable shares.
- Implication for reform design:
  - Complementary environmental policies and pro-climate regulation (higher EPS) are important to ensure that product-market reforms in energy sectors translate into improved environmental outcomes and sustainable innovation.
- Policy recommendation summary:
  - Structural reforms should aim not just for market efficiency but also for green growth.
  - Decoupling economic growth from GHG emissions is possible through comprehensive reforms and policies aimed at shifting the energy matrix away from fossil fuels.
  - Transitioning energy supply to low-carbon sources is critical.
  - Achieving environmentally sustainable growth also depends on greater efficiency in the distribution and use of energy.

### Selected evidence from appendices (examples of identified reforms)
- Australia, Electricity, 1996: Market access and structure reform to create a competitive market for bulk electricity in southern and eastern Australia from 1 July 1995. (References to pg. 68, 1995; pg. 76, 1998)
- Finland, Electricity, 2000: Privatization mandates broadened—government authorized to reduce ownership to 50.1 per cent in Altia Group and Vapo, to 20 per cent in Rautaruukki, to 15 per cent in Kemira Group, to 10 per cent in Outokumpu and to zero in Inspecta. (pg. 151, 2002; pg. 64, 2002)
- Belgium, Gas, 1996: Initial public offer (16.60 percent of capital) of the Belgian gas treatment, transmission and storage monopoly Distrigaz; government later sold its remaining share but retained one golden share.
- Slovak Republic, Gas, 2007: Wide-ranging reforms to introduce competition in energy markets; managerial separation in gas and electricity transport networks; legal unbundling virtually complete. (pg. 105, 2007)

*Source: wpiea2023120-print-pdf - Section V offers concluding remarks with policy implications.*

### Section V offers concluding remarks with policy implications.

### wpiea2023120-print-pdf - Section V offers concluding remarks with policy implications

### Data and sample
- Panel dataset of annual observations covering 25 countries over the period 1970–2020 drawn from the OECD.
- Countries in the sample include: Australia, Austria, Belgium, Canada, Czech Republic, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, Korea, Luxembourg, the Netherlands, New Zealand, Norway, Portugal, the Slovak Republic, Spain, Sweden, Switzerland, the United Kingdom, and the United States.
- Dependent variables:
  - Environmental outcomes: (i) CO2 emissions in metric tons per capita, (ii) GHG emissions in metric tons per capita, (iii) GHG emissions per unit of GDP.
  - Green growth indicators: (i) share of renewable energy supply, (ii) number of environment-related inventions per capita, (iii) number of patents for environment-related technologies per capita.
- Main explanatory variables:
  - Structural reforms in the energy sector (electricity and gas) from a narrative database of major policy changes in product market regulation (Duval et al. (2018) updated by Wiese et al. (2023) until 2020).
  - Reform coding: 0 in non-reform years, 1 in reform years (liberalizing/deregulating), and -1 in counter-reform years (tightening/regulating).
- Table 1 summary (1970-2020):
  - Product market reforms: Number of reforms 235; Number of counter-reforms 4; Reforms (% of total) 98.3; Counter-reforms (% of total) 1.7.
  - Electricity sector: Number of reforms 48; Number of counter-reforms 0; Reforms (% of total) 100.0; Counter-reforms (% of total) 0.0.
  - Gas sector: Number of reforms 49; Number of counter-reforms 0; Reforms (% of total) 100.0; Counter-reforms (% of total) 0.0.
- Environmental Policy Stringency (EPS) Index:
  - OECD EPS index scaled from 0 (not stringent at all) to 6 (very stringent); covers 28 OECD countries and 6 emerging market economies over 1990–2020.
  - EPS used to test state-dependent impacts of energy reforms.

### Econometric methodology
- Estimation approach:
  - Local Projections (LP) method to estimate impulse response functions (IRFs) of environmental outcomes and green growth to structural reforms in electricity and gas.
  - Rationale: LP is flexible for panel data with fixed effects, economizes parameters relative to VARs, and is well suited to nonlinearities and state-dependent analysis.
- Baseline specification (cumulative responses over a five-year horizon):
  - Equation (1): y_{t+k,i} − y_{t-1,i} = α_i + τ_t + β_k SR_{i,t} + θ X_{i,t} + ε_{i,t}
    - y denotes the environmental or green-growth proxy (as listed above).
    - α_i and τ_t are country and time fixed effects.
    - β_k denotes the cumulative response k years after reform; SR_{i,t} denotes structural reforms in electricity and gas.
    - Control variables: 2 lags of the structural reform shock, 2 lags of real GDP growth, inflation, and the dependent variable (lag length chosen by Akaike’s information criterion).
  - Estimation via Ordinary Least Squares (OLS).
  - Standard errors: Spatial Correlation Consistent (SCC) Driscoll-Kraay (1998) standard errors; also report 90 percent and 68 percent confidence bands.
- State-dependent (nonlinear) specification:
  - Equation (2): y_{t+k,i} − y_{t-1,i} = α_i + τ_t + β_k^L F(z_{i,t}) SR_{i,t} + β_k^H (1−F(z_{i,t})) SR_{i,t} + θ X_{i,t} + ε_{i,t}
  - Weighting function F(z) = exp(γ z) / (1 + exp(γ z)), γ>0; z is the EPS index normalized to zero mean and unit variance.
  - Interpretation: β_k^L captures impact in low-EPS states (F(z)≈1 as z→−∞); β_k^H captures impact in high-EPS states (1−F(z)≈1 as z→+∞).
  - Parameter choice: γ = 1.5 (results robust for γ between 1 and 4).
- Robustness and identification:
  - Use of Driscoll-Kraay (1998) SCC standard errors to handle heteroskedasticity, autocorrelation, and cross-country correlations.
  - Robustness checks include excluding country fixed effects, adding country-specific time trends, controlling for reforms in additional policy areas, and instrumental variable (IV) estimation.
  - IV strategy: political-economy instruments from the World Bank Database of Political Institutions:
    - Parties (discrete: left=3, center=2, right=1), System (discrete: parliamentary=2, assembly-elected=1, presidential=0), Fragmentation (continuous 0–1), Democ (Polity IV index normalized 0–1).
  - Instrument validity assessed using Kleibergen-Paap and Hansen statistics; reported tests generally reject underidentification and do not reject instrument validity.

### Main empirical findings
- Baseline (unconditional) results (cumulative five-year horizon):
  - Environmental outcomes:
    - Structural reforms in electricity and gas markets lead to higher CO2 and GHG emissions per capita, but these unconditional increases are statistically insignificant and surrounded by great uncertainty.
    - GHG emissions per unit of GDP declines below the initial level over the five-year period, showing signs of decoupling between economic growth and emissions.
  - Green growth:
    - Structural reforms lower the supply of renewable energy as a share of total energy supply initially; this negative effect is not persistent over the long run.
    - Structural reforms stimulate a sustained increase in the number of environment-related inventions and patents per capita beyond the initial level.
    - Note: energy-sector privatization in baseline leads to higher emissions across all measures, including GHG emissions per unit of GDP.
- Heterogeneity by reform type:
  - Similar response patterns found for “market access and structure” reforms and “public ownership” (privatization) reforms; IRFs for electricity and gas reforms separately are consistent with baseline results.
- State-dependent results (role of initial environmental policy stringency):
  - Striking differences by EPS:
    - In countries with stronger environmental regulations and pro-climate policies (high EPS), structural reforms in the energy sector deliver a significant and persistent decline in CO2 and GHG emissions per capita and in GHG emissions per unit of GDP in the first year and over the long run.
    - In low-EPS countries, emissions continue to increase following energy-sector reforms.
    - Reforms increase the share of renewable energy and development of environmental technologies in high-EPS countries, while the impact is the opposite in low-EPS countries.
  - Energy-sector privatization results in significantly better environmental outcomes and green growth in countries with stronger environmental regulations.
- Robustness exercises:
  - Excluding country fixed effects suggests the bias from the interaction of fixed effects and country-specific developments is negligible.
  - Including country-specific time trends leaves qualitative findings unchanged.
  - Controlling for reforms in additional areas yields results consistent with the baseline.
  - IV estimates using political-economy instruments broadly confirm baseline findings, suggesting endogeneity is not a serious concern.

### Policy implications and concluding remarks (Section V highlights)
- Climate challenge and required mitigation:
  - GHG emissions are projected to increase by only 7.5 percent by 2030 compared to the required reduction of 55 percent just to keep global warming below 1.5°C.
- Policy message:
  - Achieving necessary emissions reductions and promoting green growth requires a new development model that better balances income growth and environmental priorities by:
    - Modernizing the energy matrix away from fossil fuels.
    - Increasing efficiency in the distribution and use of energy.
    - Implementing structural reforms and policies designed to alter behavior throughout the economy.
- Role of energy-sector structural reforms:
  - Structural reforms in electricity and gas can contribute to climate change mitigation and promote green growth, but their environmental effectiveness depends critically on the strength of concurrent environmental policies (EPS).
  - In countries with strong environmental policy frameworks (high EPS), energy-sector reforms are more likely to lower emissions and foster renewable energy expansion and clean-technology innovation.
  - In countries with weak environmental policies (low EPS), similar reforms may lead to higher emissions and reduced renewable shares.
- Implication for reform design:
  - Complementary environmental policies and pro-climate regulation (higher EPS) are important to ensure that product-market reforms in energy sectors translate into improved environmental outcomes and sustainable innovation.

*Source: wpiea2023120-print-pdf - Section V offers concluding remarks with policy implications.*

### 2020. We also explore the possibility of nonlinear effects of structural reforms by taking into

### wpiea2023120-print-pdf - 2020. We also explore the possibility of nonlinear effects of structural reforms by taking into

### Key empirical findings
- Electricity and gas sector reforms so far have failed in bringing about a reduction in CO2 and GHG emissions per capita.
- There is some evidence that electricity and gas sector reforms are more effective in lowering GHG emissions per unit of GDP.
- Electricity and gas sector reforms are not associated with higher supply of renewable energy as a share of total energy supply.
- Electricity and gas sector reforms appear to stimulate a sustained increase in the number of environmental inventions and patents per capita over the medium term.
- Strong evidence of nonlinear effects: market-oriented electricity and gas sector reforms lead to better environmental outcomes and green growth in countries with stronger environmental regulations.

### Policy implications and recommendations
- Structural reforms should aim not just for market efficiency but also for green growth.
- Decoupling economic growth from GHG emissions is possible through comprehensive reforms and policies aimed at shifting the energy matrix away from fossil fuels.
- Transitioning energy supply to low-carbon sources is critical.
- Achieving environmentally sustainable growth also depends on greater efficiency in the distribution and use of energy.

### Selected evidence from appendices (examples of identified reforms)
- Australia, Electricity, 1996: Market access and structure reform to create a competitive market for bulk electricity in southern and eastern Australia from 1 July 1995. (References to pg. 68, 1995; pg. 76, 1998)
- Finland, Electricity, 2000: Privatization mandates broadened—government authorized to reduce ownership to 50.1 per cent in Altia Group and Vapo, to 20 per cent in Rautaruukki, to 15 per cent in Kemira Group, to 10 per cent in Outokumpu and to zero in Inspecta. (pg. 151, 2002; pg. 64, 2002)
- Belgium, Gas, 1996: Initial public offer (16.60 percent of capital) of the Belgian gas treatment, transmission and storage monopoly Distrigaz; government later sold its remaining share but retained one golden share.
- Slovak Republic, Gas, 2007: Wide-ranging reforms to introduce competition in energy markets; managerial separation in gas and electricity transport networks; legal unbundling virtually complete. (pg. 105, 2007)

### Robustness and inference notes (figure captions and methodology signals)
- Impulse-response style analysis with t=0 the year of the structural reform and t=1 the first year of impact.
- Confidence visualization: solid black lines denote responses to a structural reform; dark grey area denotes 90 percent confidence bands; light gray area denotes 68 percent confidence bands. Standard errors clustered at country level.
- Robustness exercises include: excluding country fixed effects; including country-specific time trends; controlling for additional reform areas; an Instrumental Variable Approach.

*Source: wpiea2023120-print-pdf (IMF).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023120-print-pdf.pdf_