## annexch1

## Source details

**Canonical URL:** [annexch1](https://www.imf.org/-/media/files/publications/weo/2020/october/english/annexch1.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/weo/2020/october/english/annexch1.pdf.md)
- [Structured JSON version](/-/media/files/publications/weo/2020/october/english/annexch1.pdf.json)

---

### Methodology
- A panel regression relates energy dependence to a third order polynomial in (log) income per capita (gdp):
  - coal_depend_it = β0 + β1 gdp_it + β2 (gdp_it)^2 + β3 (gdp_it)^3 + β4 X_i + λ_t + ε_it
  - X_i includes: share of manufacturing in nominal value added (in deviation from global average), coal reserves per capita (in logs), and hydropower potential (in logs).
  - λ_t are time fixed effects; i and t index countries and years.
- Two measures of coal dependence are used:
  - relative coal dependence: share of coal in total primary energy supply.
  - absolute coal dependence: (log of) coal consumption per capita.
- Data sources (as reported): International Energy Agency for coal measures; revised Maddison Project Database for real GDP per capita (in 2011 USD); UN national accounts database for manufacturing share; World Bank’s Development Indicators for hydropower potential; OECD for electricity market regulation index and environmental policy stringency (EPS); World Bank Climate Change Knowledge Portal for average summer and winter temperature.

### Empirical findings: income and coal dependence
- Relative coal dependence (coal share in energy mix):
  - Strong evidence of an inverse U-shaped relationship between income and the share of coal in the energy mix.
  - Coal attains its maximum share at an income level of $9600 per capita.
  - Specification (1) predicts that, between 1971 and 2017, income per capita contributed to:
    - reductions in the coal share of 6.4 percentage points in the United States.
    - reductions in the coal share of 5.2 percentage points in Japan.
    - increases in the coal share of 12.2 percentage points in India.
    - increases in the coal share of 11.3 percentage points in China.
- Absolute coal dependence (coal consumption per capita):
  - The relationship with income is highly nonlinear and S-shaped:
    - At low-income levels, coal consumption growth accelerates.
    - It reaches its maximum at middle-income level.
    - It then levels off and declines beyond a turning point.
  - The turning point after which coal consumption declines ranges from $35000 to $39000.
  - A 10 percent increase in average annual winter temperature reduces coal consumption per capita by 4 percent.
  - Warm winters reduce both relative and absolute coal dependence.

### Role of structural and policy controls
- Manufacturing sector:
  - A larger manufacturing sector modestly increases coal consumption because manufacturing is coal intensive.
  - The decline of US manufacturing between 1971 and 2018 contributed to a modest reduction in the US coal share by 1.2 percentage points.
  - The rise of China manufacturing over the same period contributed to an increase in China’s coal share by 2.1 percentage points.
- Electricity market deregulation and pollution limits:
  - A one standard deviation increase in electricity market deregulation lowers the coal share by 0.59 percentage point.
  - A one standard deviation increase in limits on pollution lowers the coal share by 0.65 percentage point.
- Energy endowments and weather:
  - An increase of hydropower potential of one standard deviation from the mean reduces the share of coal by 4.4 percent.
  - A one standard deviation increase in coal reserves per capita increases the coal share by 11.1 percentage points.
  - Examples:
    - Norway would increase its coal share by 11.3 percentage points if it had an average hydropower potential.
    - Reductions in coal share of 3.5 percentage points in India, 7.0 percentage points in China, and 15.5 percentage points in the United States would be achieved if they had an average coal reserve per capita.
  - Weather effect: a country experiencing the lowest instead of the highest possible average annual winter temperature since 1971 will see its share of coal increase by 5 percentage points.

### Coal production, trade, and macro-financial risks
- 2017 global coal statistics:
  - Total coal production: about 3800 million metric tons.
  - Equivalent value: $506 billion.
  - Equivalent share of global GDP: 0.63 percent.
  - For comparison: the oil expenditure share is about 3 percent.
- Coal exports and GDP exposure:
  - Coal exports represent about 3 percent of GDP in Mozambique and Australia.
  - Coal exports represent about 1 percent of GDP in South Africa, Colombia, and Indonesia.
- Reserves and stranded asset risk:
  - The value of coal reserves is multiples of GDP in various countries, making the risk of stranded coal assets macro relevant for some major coal-producing countries.
  - The needs of major coal consumers are typically met domestically, which raises political economy hurdles for policies curbing domestic coal consumption because moving away from coal lowers demand for domestically mined product and can lead to losses for the local mining industry and workers.
  - In many European countries, coal imports have displaced domestic coal production in recent decades, meaning a large portion of losses from a coal phaseout would not be borne by domestic citizens.
- Historical adjustment example (United States):
  - Rapid transition from coal to natural gas driven by the shale boom led to:
    - decline in coal mine employment.
    - a record number of bankruptcies among coal mining firms.
    - a sharp decline in the Dow Jones US Coal Index.
  - Mining employment fell from almost 90,000 to about 50,000 from 2012 to 2016 (US Bureau of Labor Statistics).
- Top producers and shares (2017, as reported):
  - The top three coal-producing countries (China, United States, India) accounted for about 70 percent of world production and 78 percent of world consumption.
  - China alone was responsible for [47.3] percent of world production and [51.5] percent of world consumption, respectively.

### India: dependence on coal, drivers, and transition implications
- India’s role and emissions:
  - India’s per capita emissions are about one-tenth that of the United States.
  - In 2018 India accounted for 6.4 percent of global energy consumption and 6.8 percent of global carbon dioxide emissions.
  - India’s share of emissions is expected to exceed 10 percent by 2030 (IEA 2018).
- Drivers of high coal reliance in India:
  - About 70 percent of the coal consumed in India is produced domestically; India is the second-largest hard coal producer after China.
  - Coal is affordable, available, and reliable compared with alternatives:
    - Natural gas would need to be imported and requires huge infrastructure investment in gasification facilities and pipelines.
    - Renewable energy has low operating costs once installed but requires sizable upfront investment (per megawatt) and has a high import content.
  - Political economy factors:
    - Employment in mining and its footprint on regional economies is non-negligible.
    - Coal accounts for about 44 percent of Indian Railways’ revenue and half of its profit, allowing the railway operator to cross-subsidize passenger fares.
    - Some coal-producing states (Jharkhand, Odisha, West Bengal, Bihar, Chhattisgarh, Telangana, and Madhya Pradesh) use coal levies to partly finance spending.
    - Electricity prices for households are low and controlled, limiting scope to raise the modest national coal levy.
    - Central government control of large utility firms and coal producers, and long-term purchasing agreements between utilities and coal producers, can disincentivize faster green energy expansion.
  - Lack of natural gas infrastructure limits both adoption of gas-fired plants and the replacement of coal with renewables because there is insufficient backup generation capacity to compensate renewables intermittency.
- Policy implications and recommendations for India:
  - Build on current policies that have successfully encouraged investment in renewables.
  - Support development of natural gas infrastructure.
  - Concessionary financing from advanced economies would help stimulate India’s green policy adoption and its fairness.
  - The green transition would reduce financial stability risks associated with Indian banks’ exposure to coal assets in the power and mining sector, which could become stranded by an international carbon tax or local environmental regulations.
  - Offer compensation to the mining and power sectors (possibly paid for by raising carbon tax revenue) for the early retirement of old and inefficient coal-fired power plants to reduce financial stability risks associated with stranded assets during a rapid transition.

*Source: annexch1*

---


_Source: https://www.imf.org/-/media/files/publications/weo/2020/october/english/annexch1.pdf_