Global electricity demand fell slightly (-0.1%) in 2020, the first fall since 2009. But this pause has already ended: by December 2020, electricity demand was already higher than in December 2019 (India +5%, EU +2%, Japan +3%, South Korea +2%, Turkey +3%, US +2%).
Wind and solar generation rose robustly in 2020 by 15% (+314 TWh). This meant that wind and solar produced almost a tenth (9.4%) of the world’s electricity last year, doubling from 4.6% in 2015. Many G20 countries now get around a tenth of their electricity from wind and solar: India (9%), China (9.5%), Japan (10%), Brazil (11%), the US (12%) and Turkey (12%). Europe is leading the way, with Germany at 33% and the United Kingdom at 29%. Indonesia, Russia and Saudi Arabia still have near-zero.
Coal fell a record 4% (-346 TWh). This was similar to the rise in wind and solar power of 314 TWh, more than the UK’s entire electricity production. This dwarfed the aggregate changes across global electricity: demand fell 23 TWh, gas and oil fell 12 TWh. A rise in hydro of 94 TWh was mostly countered by a fall of 104 TWh of nuclear. In comparison, coal collapsed almost everywhere, with large falls in the US (-20%), EU (-20%) and even India (-5%).
China’s coal generation rose by 2% in 2020. That was because electricity demand growth continued to outstrip new clean electricity. China’s electricity demand was 33% higher in 2020 than in 2015, rising by more than all electricity demand in India in 2020. Across those five years, China’s fossil-free generation met only 54% of the rise in electricity demand, so 46% was met from fossil generation. That pushed China’s coal generation 19% higher in five years. China is now responsible for more than half (53%) of the world’s coal-fired electricity, up from 44% in 2015.
Electricity demand rose 11% (+2536 TWh) since 2015, but the increase in clean electricity generation (+2107 TWh) didn’t keep up. That led to an increase in overall fossil generation: gas-fired electricity rose 11% (+562 TWh) and coal fell only 0.8% (-71 TWh). As a result, power sector CO2 emissions were around 2% higher in 2020 than in 2015. Fossil-free electricity met only 54% of the rise in electricity demand in China, 57% in India and 37% in Indonesia. Meanwhile in Europe, and especially the US, coal’s fall was caused not only by a rise in clean electricity, but also a rise in gas generation. Of the 10% rise in global gas generation since 2015, half of that was in the United States.
Progress is nowhere near fast enough. Despite coal’s record drop during the pandemic, it still fell short of what is needed. Coal power needs to collapse by 80% by 2030 to avoid dangerous levels of warming above 1.5 degrees. We need to build enough clean electricity to simultaneously replace coal and electrify the global economy. World leaders have yet to wake up to the enormity of the challenge.
Despite some progress, China is still struggling to curb its coal generation growth. Fast-rising demand for electricity is driving up coal power and emissions. More sustainable demand growth will enable China to phase out its large coal fleet, especially the least efficient sub-critical coal units, and provide greater opportunity for the country to attain its climate aspirations.
India has started its clean electricity transition. India now needs to ramp up wind and solar considerably in the next decade to both replace coal and meet rising electricity demand. India has the opportunity to ensure that coal generation doesn’t see a resurgence after the last two years of coal falling.
Download the full report and data
This annual report analyses electricity data from every country in the world to give the first accurate view of the global electricity transition in 2020. It aggregates generation data by fuel by country from 2000. 68 countries comprising 90% of world electricity generation have full-year data to 2020 and have formed the basis of an estimate for changes in worldwide generation. All remaining countries have full data as far as 2019. G20 countries, which comprise 84% of world electricity generation, each have a separate in-depth country analysis.
Generation data is mapped into nine generation types. More information on mapping for different sources and countries can be viewed below. For the purpose of analysis these generation types are aggregated into different groupings as follows:
1 Solar includes both solar thermal and solar photovoltaic generation, and where possible distributed solar generation is included.
2 Where possible, hydro generation excludes any contribution from pumped hydro generation.
3 Bioenergy generation includes generation from combustible renewables. For certain historical data sources all waste generation (renewable and non-renewable) is not disaggregated from other combustible renewables generation – this has all been mapped to bioenergy.
4 Other renewables generation includes geothermal, tidal and wave generation.
5 Other fossil generation includes generation from oil and petroleum products, as well as manufactured gases.
For the purposes of this report, renewables are classified in line with the IPCC and include bioenergy. However, the climate impact of bioenergy is highly dependent on the feedstock, how it was sourced and what would have happened had the feedstock not been burnt for energy. The current EU bioenergy sustainability criteria do not sufficiently regulate out high-risk feedstocks and therefore electricity generation from bioenergy cannot be automatically assumed to deliver similar climate benefits to other renewables sources (such as wind and solar) over timescales relevant to meeting the commitments of the Paris Agreement. For more information please see Ember’s reports: The Burning Issue (June 2020) and Playing with Fire (December 2019).
Data from the United States Energy Information Administration’s (EIA) international data browser forms the backbone of this report. With the exception of China, India, EU-27 and the United States, all data from 2000-2019 is taken from this source. For some generation types in non-OECD countries, 2019 data had not yet been published by the EIA. Where available, we used national data to estimate these values, otherwise, the average change for that generation type seen over 2015-2018 was added onto the 2018 generation value.
EIA international data does not disaggregate generation from fossil fuels. This was performed by Ember using two methods. If possible, the split between fossil fuels was estimated using the ratios of fossil generation types in BP’s statistical review of world energy. Fossil generation was disaggregated for any remaining countries by using the capacity split by fossil fuel type, taken from the WRI’s global power plant database.
Data for 2020 is estimated using national sources. The year-on-year changes in generation for each fuel type from these sources are added onto the historical 2019 data to obtain a value for 2020. Data sources for Argentina, Canada, Ecuador, Kazakhstan and Russia provide no disaggregation of thermal generation – for these countries, estimates of the split between different generation types were made using the IEA’s monthly electricity generation statistics and BP’s statistical review of world energy.
World data for 2000-2019 is the sum of all country data. World data for 2020 is estimated by summing generation for all countries where we have 2019 and 2020 data. Together this comprises 90% of global generation. The percentage changes in each fuel type from 2019 to 2020 are then applied to the world generation data for 2019, to create an estimate of global generation in 2020.
Report design by Designers for Climate
Thank you to all peer reviewers
Ember is an energy think tank that aims to accelerate the clean energy transition with data and policy. Ember is the trading name of Sandbag Climate Campaign CIC, a Community Interest Company registered in England & Wales #06714443. 'Ember' and 'Sandbag' are trademarks held at the United Kingdom and European Union Intellectual Property Offices. All content is released under a Creative Commons Attribution Licence (CC-BY-4.0). Website powered by 100% renewable electricity.